Skip to main content

Full text of "Recommendations and Reports - 1989"

See other formats





Office of the Chairman 

Administrative Conference of the United States 



Administrative 
Conference 
of the 
United States 



Recommendations 
and Reports 



1989 



Administrative Conference of the United States 



The Administrative Conference of the United States was established 
by statute as an independent agency of the federal government in 
1964. Its purpose is to produce improvements in the efficiency, 
adequacy, and fairness of procedures by which federal agencies 
conduct regulatory programs, administer grants and benefits, and 
perform related governmental functions. 

To this end, the Conference conducts research and issues reports 
concerning various aspects of the administrative process and, when 
warranted, makes recommendations to the President, Congress, par- 
ticular departments and agencies, and the judiciary concerning the 
need for procedural reforms. Implementation of Conference recom- 
mendations may be accomplished through direct action on the pari of 
the affected agencies or legislative changes. 



Office of the Chairman 

Administrative Conference of the United States 



Administrative 
Conference 
of the 
United States 



Recommendations 
and Reports 



1989 



Cite as: 
1989 ACUS 



CONTENTS 



RECOMMENDATIONS and STATEMENTS of the 

Administrative Conference of the United States 

Recommendation 89-1: Peer Review and Sanctions in the 

Medicare Program 3 

Recommendation 89-2: Contracting Officers' Management 

of Disputes 12 

Recommendation 89-3: Conflict-of-interest Requirements 

for Federal Advisory Committees ... 1 7 
Recommendation 89-4: Asylum Adjudication Procedures ... 22 
Recommendation 89-5: Achieving Judicial Acceptance of 

Agency Statutory Interpretations ...31 
Recommendation 89-6: Public Financial Disclosure by 

Executive Branch Officials 34 

Recommendation 89-7: Federal Regulation of 

Biotechnology 42 

Recommendation 89-8: Agency Practices and Procedures for 

the Indexing and Public Availability 

of Adjudicatory Decisions 47 

Recommendation 89-9: Processing and Review of Visa 

Denials 51 

Recommendation 89- 1 0: Improved Use of Medical Personnel 

in Social Security Disability 

Determinations 54 

Statement 14: Statement on Mass Decisionmaking 

Programs: The Alien Legalization 

Experience 60 

Statement 15: Procedures for Resolving Federal Personnel 

Disputes 66 



REPORTS for Recommendations and Statements 
Rec. 89-1: Timothy Stolzfus Jost. Administrative Law 

Issues Involving the Medicare Utilization and 

Quality Control Peer Review Organization 

(PRO) Program: Analysis and 

Recommendations 77 

Rec. 89-2: Richard J. Bednar. Government Contracting 

Ojjvcers Should Make Greater Use of ADR 

Techniques in Resolving Contract Disputes . . .149 
Rec. 89-3: Richard K. Berg. Conflict-of-interest for Members 

of Federal Advisory Commitiees 205 



Rec. 89-4: David A. Martin. Reforming Asylum 

Adjudication: On Navigating the Coast of 

Bohemia 233 

Rec. 89-5: Robert Anthony. Which Agency Interpretations 

Should Bind Citizens and the Courts? 369 

Rec. 89-6: Thomas D. Morgan. Public Financial Disclosure 

by Federal Officials: A Functional 

Approach 447 

Rec. 89-7: Sidney A. Shapiro. Biotechnology and the 

Design of Regulation 475 

Rec. 89-8: Margaret Gilhooley. The Availability of Decisions 

and Precedents in Agency Adjudications: The 

Impact of the Freedom of Information Act 

Publication Requirements 545 

Rec. 89-9: James A. R. Nafziger. Report on Reviewability of 

Visa Denials by Consular Officers 587 

Rec. 89-10: Frank S. Bloch. The Use of Medical Personnel in 

Social Security Disability Determinations . . . . 73 1 
Stmt. 14: David S. North and Anna Mary Portz. Decision 

Factories: The Role of the Regional Processing 

Facilities in the Alien Legalization 

Programs 819 

Stmt. 15: William V. Luneburg. The Federal Personnel 

Complaint Appeal and Grievance Systems: A 

Structural Overview and Proposed 

Revisions 395 



RECOMMENDATIONS 

AND 

STATEMENTS 



Digitized by the Internet Archive 

in 2010 with funding from 

Public.Resource.org and Law.Gov 



http://www.archive.org/details/adminconf198900unse 



Recommendation 89-1 

Peer Review and Sanctions in tlie Medicare 
Program 



As the Administrative Conference noted in Recommendation 
86-51, the Medicare program relies heavily on implementation of 
federal requirements by localized carriers, intermediaries and, 
increasingly, peer review organizations (PROs). 

The PRO system was created in 1982. It is made up of state- 
wide, physician-controlled organizations under individual 
contracts with the Department of Health and Human Services 
(HHS). These contracts are negotiated pursuant to a general 
contractual "Scope- of -Work" promulgated by HHS every three 
years. PROs are delegated a number of important 
responsibilities under the Medicare system. They identify 
substandard, unnecessary or inappropriate services rendered to 
Medicare beneficiaries, and oversee education and corrective 
actions for substandard providers (e.g., hospitals) and medical 
practitioners. They also recommend to HHS that it sanction 
providers and practitioners when they find seriously improper 
practices, deny Medicare payment for inappropriate or 
unnecessary services, and protect the rights of beneficiaries. 

This recommendation follows the suggestion made in 
Recommendation 86-5 that the PRO program was deserving of 
further study. It recognizes the evolutionary nature of the PRO'S 
role in Medicare, and the administrative difficulties posed for 
HHS in overseeing this decentralized program -- especially since 
new legislative directions affecting the program appear regularly, 
often contained in year-end omnibus budget reconciliation acts. 
Nevertheless, the Conference urges the Department (and, where 
necessary. Congress) to make changes designed to improve the 
accessibility of PRO-related policies, the fairness and firmness of 
PRO sanctions imposed on providers and practitioners, and the 
effectiveness of PRO safeguards for beneficiary rights. 

In Paragraph A of the Recommendation, the Conference 
urges several enhancements of HHS* current practices in 
disseminating, making accessible, and soliciting comments on, 
PRO program guidelines of general applicability, including the 



IaCUS Recommendation 86-5, Medicare Appeals, 1 C.F.R 305.86-5. 



4 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

scopes of work, manuals, and the criteria and norms used to 
evaluate medical care. Paragraph B seeks to promote 
Improvements in the PRO'S assigned duty of investigating 
complaints by beneficiaries, and urges Congress to allow PROs 
to act in response to oral complaints. 

Paragraph C recommends invigorating the process of 
investigating and adjudicating sanctions against health care 
practitioners and providers charged with violations of their 
obligations under the Medicare program. The current sanction 
process begins when a PRO gives formal notice to the 
practitioner or provider involved that it considers that poor 
quality care may have been rendered or that other violations 
have occurred. The PRO is required to have at least one quite 
formalized meeting with the practitioner or provider to discuss 
the allegations that the care rendered either "failed in a 
substantial number of cases substantially to comply" with the 
statutory obligation to render proper medical care, or "grossly 
and flagrantly violated such obligations in one or more 
instances." 42 U.S.C. 1320c-5(b). {In the former type of case, 
at least two meetings are required.) If, after the meeting, the 
PRO believes that violations have occurred, it recommends to 
the HHS Office of Inspector General (OIG) that a sanction be 
imposed, either in the form of an exclusion from participation in 
the Medicare program for some period of time, or a civil 
monetary penalty of no more than the amount of the cost of 
medically improper or unnecessary services. If the OIG agrees 
that violations have occurred, and in addition finds that the 
practitioner or provider is unwilling or unable to comply with 
the obligations to render proper care, the OIG may impose one 
of these sanctions. If the sanction is exclusion, it becomes 
effective fifteen days after notice. 2 The sanction is appealable to 
an ALJ, then to the Appeals Council; judicial review is 
subsequently available. 

This recommendation seeks to balance the vital interest in 
protecting the health and safety of program beneficiaries and 
the need to assure fairness to the accused provider or 
practitioner whose livelihood is at stake and whose services 
might be needed. The Conference urges that the current PRO 
sanction process be streamlined. It also urges that all providers 
and practitioners, not just some, be permitted to seek a stay of 



^Certain practitioners in rural areas are permitted to have the exclusion 
stayed, pending OIG proof that the practitioner would pose a "serious risk" to 
program beneficiaries If allowed to remain in the program during the pendency 
of the administrative appeal. 



RECOMMENDATION 89-1, MEDICARE SANCTIONS PROCESS 5 

an HHS order to exclude them from the Medicare program, in a 
proceeding akin to that of a temporary restraining order at the 
administrative law judge adjudication stage of that process. 
However, the burden would be on the practitioner or provider to 
show that no serious risk would be posed to beneficiaries during 
the pendency of the administrative appeal. The Conference also 
urges legislative changes that, while maintaining the 
requirement that the OIG prove that violations have occurred, 
would eliminate the additional requirement of proving that the 
practitioner or provider is unwilling or unable to comply with 
the obligations to provide quality care. The offenses or 
oversights, which have been found both by peers (PROs) and 
regulators (OIG) to be substantial or gross and flagrant, already 
serve as indicators of inability or unwillingness to comply. 
Under the current law, before excluding a provider or 
practitioner on the basis ^f these findings, the government must 
bear an additional evidentiary burden that is inappropriate for 
this type of proceeding. It must prove what amounts to a 
speculative negative -- that violators would be unwilling or 
unable to comply with the law in the future. The apparent 
result of this evidentiary requirement has been to chill the 
initiation of exclusion proceedings against providers and 
practitioners who are providing improper care or otherwise 
violating the law. Further, the Conference recommends 
legislative changes to provide for meaningful civil money 
penalties, as well as for the current sanction of excluding 
providers and practitioners from the program. It should be 
noted that the Conference views the changes in the sanction 
procedure contained in this paragraph as a unified package, one 
that in its present form balances conflicting interests but that 
will become unbalanced if any one significant portion were not 
to be accepted. 

Paragraph D urges changes In the PRO statute and 
regulations to ensure that beneficiaries are better Informed of 
their rights to appeal decisions concerning their lack of coverage 
or discharge from a hospital or other facility, and that they will 
not be discharged until such appeals are resolved. Paragraph E 
covers the PRO'S role in denials of payment for care determined 
to be unnecessary, substandard or rendered in an inappropriate 
setting. It recommends that HHS implement in final rules 1985 
legislation concerning PRO denials for substandard care. 3 It 
also urges HHS to amend its rules to require that PROs not 



30n Januajy 18, 1989, HHS published a proposed rule covering this 
subject. 54 Fed. Reg. 1956. 



6 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

make any final decisions affecting pajnnent without adequate 
review by medical practitioners who are qualified in the relevant 
area. Finally, Paragraph F urges HHS to take steps to permit 
PROS to share information with provider facilities and state 
medical boards. 



RECOMMENDATION 



A. Publication and Dissemination of PRO Program Guidelines 

1. HHS should enhance its current practice of 
publishing and disseminating all Peer Review Organization 
(PRO) program rules having a substantial effect on 
providers, medical practitioners and beneficiaries by taking 
the following steps: 

(a) Notlce-and-comment procedures should be used 
for rulemaking except when the agency for good cause 
finds that notice and public procedure thereon are 
impracticable, unnecessary, or contrary to the public 
Interest. 4 

(b) Proposed PRO "scopes of work" and any 
generally applicable modifications or Interpretations of 
the responsibilities of PROs during a contract cycle 
should be published in the Federal Register and 
disseminated to relevant Interest groups. Interested 
parties should be allowed 30-45 days for commenting, 
unless explicit Congressional deadlines would be 
contravened thereby, or unless there is good cause for 
Immediate Implementation. 

(c) HHS should make PRO contracts, manual 
instructions, and other guidelines of general 
applicability regarding the PRO program readily 
available to the public at convenient locations. 
Including social security offices. HHS should publish 
an updated list of such materials in the Federal 
Register at least quarterly. 

2. HHS should encourage PROs to use outreach and 
consensus-buUdlng techniques analogous to negotiated 
rulemaking when they are developing criteria and norms for 



^See ACUS Recommendation 83-2, The "Good Cause" Ebcemption from APA 
Rulemaking Requirements, 1 C.F.R 305.83-2. 



RECOMMENDATION 89-1, MEDICARE SANCTIONS PROCESS 7 

PRO review of the quality, necessity and appropriateness of 
medical care. 5 HHS should further encourage PROs to 
make these criteria and norms consistent nationwide. 

B. PRO Investigations of Beneficiary Complaints 

1. Congress and HHS should coordinate the system of 
PRO review of beneficiary complaints concerning quality of 
services with other federal and state regulatory schemes. 
Initially, priority consideration should be given to complaint 
investigations in the hospital setting, where PROs have the 
most expertise and where alternative means to investigate 
complaints are least available. 

2. Congress should amend 42 U.S.C. 1320c-3(a)(14) 
to permit PROs to investigate and otherwise act on oral 
complaints concerning the quality of services. Until it does 
so, HHS should require PROs to receive such oral 
complaints from beneficiaries or witnesses, and reduce 
them to writing, before acting on them. 

3. HHS should require PROs to use investigative 
techniques that, so far as may be feasible, protect from 
disclosure the identity of complainants who do not 
expressly and voluntarily consent to such disclosure. 
Where the Identity of a complainant who desires anonymity 
cannot be kept confidential, the PRO should give the 
complainant the option of withdrawing the complaint in lieu 
of disclosure, although the PRO may at its discretion 
continue to investigate the underlying problem. 

4. HHS should amend the PRO Scope of Work to 
conform to the 1986 Omnibus Budget Reconciliation Act by 
requiring PROs to Inform beneficiaries fully regarding the 
final disposition of all complaints, whether involving 
providers or practitioners. PROs also should be required 
promptly to Inform providers and practitioners of the final 
disposition of investigations involving them. 

5. HHS should establish guidelines and a significantly 
more expedited schedule than the current several-month 
process for PROs to complete initial Investigations of 
complaints of potentially life-threatening quality 
deficiencies. HHS also should establish procedures for 
receiving and acting on requests for intervention in cases 
where PROs do not process complaints on a timely basis. 



5See ACUS RecommendaUons 82-4, 85-5, Procedures for Negotiating 
Proposed Regulations. 1 C.F.R 305.82-4, 85-5. 



8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

C. Sanctions Against Providers or Practitioners Who Have 
Provided Improper or Unnecessary Services 

Congress should streamline the sanction process by taking 
the following interrelated steps to promote heightened 
enforcement, while preserving fairness to the accused provider 
or practitioner: 

1. HHS should seek to ensure greater uniformity 
among PROs through training and the development of a 
model sanction referral form. To preserve needed 
healthcare resources, HHS and the PROs should continue 
to emphasize education and corrective action rather than 
sanctions as the primary means of addressing quality 
problems. HHS should also amend its rules (a) to require 
that, once a PRO determines that there is a quality problem 
for which a sanction is the appropriate intervention, it 
immediately start +the sanction process, and (b) to provide 
that, ordinarily, there will be only one formal meeting 
between the PRO and the accused provider or practitioner 
after the sanction proceeding has been initiated. 

2. Congress should amend the PRO statute to offer all 
providers and practitioners (urban and rural), upon their 
receipt of an HHS notice of exclusion pursuant to 42 U.S. C. 

1320c-5(b), the opportunity for a preliminary hearing and 
decision. Such a proceeding would be conducted by an ALJ 
on the issue of whether the provider or practitioner would 
pose a serious risk to patients during the pendency of the 
subsequent ALJ proceeding on the merits of the exclusion. 
The preliminary hearing would be in the nature of a 
temporary restraining order proceeding, and would arise 
and be conducted according to the following procedures: 

(a) If, within 10 days of receipt of notice of the 
exclusion, the provider or practitioner appeals the 
decision of the HHS Office of Inspector General (OIG) 
imposing an exclusion, a preliminary hearing on the 
"serious risk" issue should take place before the 
exclusion takes effect. 

(b) If the provider or practitioner establishes at the 
preliminary hearing that continued participation in the 
Medicare program pending the ALJ's decision on the 
underlying appeal will not pose a serious risk to 
patients, or that such participation can be restricted to 
preclude such risk, the HHS exclusion order shall be 
stayed or modified by the ALJ until the ALJ issues a 
final decision on the merits of the exclusion. 



RECOMMENDATION 89-1, MEDICARE SANCTIONS PROCESS 9 

(c) The ALJ must render the preliminary decision 
on the "serious risk" issue as quickly as possible but 
within no more than 30 days after the filing of the 
appeal, and a final decision on the exclusion within a 
time period reflecting assignment of the highest priority 
to the adjudication. 

3. Congress should retain the requirement in 42 U.S. C. 
132C)c-5(b)(l) that sanctions be based on determinations 

that a practitioner or provider has either (A) "failed in a 
substantial number of cases substantially to comply" with 
statutory obligations to render appropriate and quality care, 
or (B) "grossly and flagrantly violated such obligations in 
one or more instances." However, Congress should 
eliminate the separate and additional requirement in 42 
U.S.C. 1320c-5{b)(l) that the OIG must determine the 
provider's or practitioner's "unwillingness or lack of ability 
substantially to comply" with program obligations before 
imposing sanctions on the provider or practitioner. 

4. Currently the PRO statute [42 U.S.C. 1320c-5(b)(3)l 
limits monetary penalties to "the actual or estimated cost of 
. . . medically improper or unnecessary services." In order 
to provide for a wider range of sanctions, Congress should 
amend the PRO statute to allow the OIG to assess a 
substantial civfl money penalty for each violation against 
providers and practitioners who are found to have grossly 
and flagrantly violated their obligations on one or more 
occasions, or to have substantially violated such obligations 
in a substantial number of cases. The OIG should be given 
the discretion to impose such monetary penalties In 
addition to an exclusion where appropriate. 

5. HHS should assign PRO sanction cases to ALJs 
attached to the Departmental Appeals Board (who currently 
hear other sanction cases in the Department) rather than to 
Social Security ALJs, as is the current practice. 

D. Notice to Beneficiaries of Noncoverage 

1. Congress should amend 42 U.S.C. 1320c-3(e)(3) to 
assure that hospitalized beneficiaries who appeal the 
hospital's notice of noncoverage by noon of the day 
following receipt of the notice, should not have such 
coverage discontinued untU the PRO rules on their request 
for review. 

2. HHS should amend the PRO regulations to assure 
that, at the time a hospital informs beneficiaries of its 



10 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

decision to discharge them or of the discontinuance of 
coverage, they are informed of their discharge appeal rights 
under the PRO program. 

3. The notice of a right to appeal should be on a form 
drafted by HHS (developed in consultation with beneficiary 
organizations and other interested parties), and should 
include a concise and easily understood statement of the 
basic beneficiary right to a no-liability appeal to the PRO. If 
the current system of separate appeal tracks (depending on 
whether the hospital and attending physician concur or not) 
is retained, separate notices should be given for each track 
to avoid the confusion caused by a notice that describes 
multiple procedures. 

E. PRO Denials of Payment for Substandard or Unnecessary 
Care 

1. HHS should proceed expeditiously to final 
rulemaking to implement PRO authority, contained in 42 
U.S.C. 1320c-3(a)(2), to deny payment to practitioners or 
providers for care that does not meet professionally 
recognized standards. 

2. HHS should require by regulation that PROs not 
make final utilization review denials (denials of payment for 
care that has been determined to be unnecessary or 
rendered in an inappropriate setting) until a proposed 
denial and the response to it by the affected provider or 
practitioner have been reviewed by at least one practitioner 
qualified by professional training and experience relevant to 
the matters in controversy. Although HHS should at a 
minimum apply the same standard to reviews of denials of 
payment for failure to meet professional standards of care, 
it may be appropriate in this context to require that the 
review be performed by a physician practicing in the same 
care specialty. 

F. PRO Sharing of Information 

1. HHS should issue PRO manual instructions and 
amend the Scope of Work in order to implement the 
Congressional mandate requiring the sharing of information 
among the PROs and state medical boards and licensing 
authorities regarding practitioners and providers who 
violate quality standards, and should modify its current 
confidentiality and disclosure regulations to require that a 
copy of any PRO final sanction recommendation be provided 



RECOMMENDATION 89-1, MEDICARE SANCTIONS PROCESS 1 1 

to such bodies. HHS should explore the feasibility of 
including sanction recommendations in the National 
Practitioner Data Bank. 

2. HHS should amend PRO regulations to require PROs 
to share with hospitals information about confirmed 
violations of quality of care standards involving doctors on 
the staffs of such hospitals, including the contents of 
corrective action plans. 



Recommendation 89-2 

Contracting Officers' Management of Disputes 



An increasing number of problems in the management of 
government contracts are now referred to lawyers, accountants, 
and judges for resolution. This accelerating trend has tended to 
deemphasize the responsibility of the agency contracting 
officers, who (in most agencies) have traditionally played a key 
role in the procurement process, including dispute handling, i 
Many contracting officers ("COs") today are subject to restrictive 
regulations and close oversight that can inhibit their willingness 
to negotiate settlements. For this and other reasons, many 
cases proceed to needless litigation that are in fact susceptible 
to prompt, direct resolution by COs at an early stage when 
parties are often less entrenched and more cognizant of program 
interests. 2 

Several Conference studies have demonstrated opportunities 
for improving agencies' resolution of contract disputes 
consonant with the Contract Disputes Act's^ goal of expeditious 
resolution without disrupting performance. ^ While a few 



1 Conference Recommendation 87-11, Alternatives for Resolving Government 
Contract Disputes, 1 CFR 306.87-11, describes one aspect: 

The dispute handling system established bv the Contract 
Disputes Act begins with the contracting officer f'CO"), an agency 
official whose function Is to enter Into and administer government 
contracts. Any claim arising out of a contract is to be presented 
to the CO. The CO has a dual role: to represent the government 
as a party to the contract, but also to make Initial decisions on 
claims suDJect to certain procedural safeguards. If the dispute is 
not amicably resolved, the CDA requires the CO to issue a brief 
written decision stating his or her reasons. A contractor 
dissatisfied with a CO's decision may appeal either to an agency 
board of contract appeals or directly to the U.S. Claims Court, 
where proceedings become considerably more formal. 

2Thls report addresses only dispute resolution during contract performance; 
it does not extend to controversies which arise during the contract formation 
process. 

341 U.S. Code 601-613; 5 U.S.C. 5108(c)(3); 28 U.S.C. 1346(a)(2), 
149(a)(2). 2401(a). 2414. 2510. 2517. 31 U.S.C. 1304(a)(3)(C) (1982); enacted 
November 1, 1978 by Pub. L. No. 95-563. 92 Stat. 2383. 

^Section 33.204 of the Federal Acquisition Regulation, which guides agency 
procurement practices, Includes the following possible inducement to ADR; 



RECOMMENDATION 89-2, CONTRACTING OFFICERS' ADR 13 

agencies have experimented with alternative means of dispute 
resolution at the appeal level, these methods are even more 
likely to be useful prior to issuance of a contracting officer 
decision. This potential has been neglected. Current training 
for COs does not address ADR and gives minimal attention to 
negotiation skills. These methods^ serve the agency by helping 
to expedite dispute handling. They serve the parties by keeping 
outcomes in the control of the contracting parties, preserving 
cooperative business relations, avoiding litigation (and the 
concomitant loss of control as to results), and— most important- 
allowing the parties to return to concentrating on productive 
work rather than conflict. 

This recommendation builds on an earlier one (87-11), in 
which the Conference focused primarily on possible uses for 
consensual means of resolving contract disputes at the appeal 
level. It identified the decreased authority of COs as a major 
factor contributing to the Inefficiency and cost of resolving many 
conflicts. Recommendation 87-1 1 (in pertinent part) calls for (1) 
legislation, an executive order, by the Office of Federal 
Procurement Policy, policy statement, and Federal Acquisition 
Regulation changes to encourage COs, before issuing a decision 
likely to be unacceptable to a claimant, to explore use of ADR to 
resolve their differences; (2) agency adoption of policies 
encouraging ADR and regular use of rules or notices to alert 
COs and other parties to ADR availability; (3) agency 
designation of an employee to serve as an ADR specialist in 
connection with contract disputes; and (4) agency attention to 
the need to offer training in negotiation and other ADR skills to 
COs and others involved in contract disputes. 

The instant recommendation seeks to go further to enhance 
the CO's ability and authority in the resolution of contract 
disputes. Calling for CO training in negotiation and dispute 
handling, as well as increased use of ADR techniques as part of 
a CO's decisionmaking process, it supplements the prior 



In appropriate circumstances, the contracting officer, before 
Issuing a decision on a claim, should consider the use of informal 
discussions between the parties by individuals who have not 
participated substantially in the matter in dispute, to aid in 
resolving the differences. 

This suggestion for a "fresh look" at the issues recognizes the potential 
usefulness of an objective evaluation. 

^They include arbitration, mediation, minitrial, factfinding, convening, 
facilitation and negotiation. These are defined in the Appendix to Conference 
Recommendation 86-3, Agencies' Use of Alternative Means of Dispute Resolution, 
1 CFR 306.86-3. 



14 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



recommendation by focusing on the Integration of consensual 
dispute resolution into already existing dispute and training 
systems at the CO level, overcoming obstacles to ADR use, and 
practical guidance in improving CO- level dispute resolution. 



RECOMMENDATION 



1. Agencies with significant acquisition activity, acting in 
consultation with expert groups, should encourage COs, and 
other key personnel involved in the resolution of contract 
disputes, to make greater efforts routinely to consider and 
utilize ADR to help resolve claims. Since dispute resolution at 
the CO level is very much a shared activity, these persons may 
include program and project managers, attorneys, auditors, 
engineers, specialists In pricing, packaging, production, 
maintenance and quality control, and other technical experts or 
contracting officials. These agencies should undertake 
comprehensive programs of promoting ADR at the CO level. The 
programs should include application of ADR techniques in 
specific test cases, conduct of training, case screening, and 
information and guidance for personnel and contractors. 

2. Agency heads should direct senior officials within the 
acquisition hierarchy to act as proponents for dispute 
resolution, with the specific mission of developing more effective 
contact dispute resolution practices. Agencies with extensive 
acquisition activity should designate a senior official within the 
acquisition hierarchy with the specific mission of developing 
more effective contract disputes resolution practices. This 
official's mission would Include challenging barriers to wider 
ADR use, educating disputants in industry and government, 
and improving understanding and use of ADR procedures at the 
CO level. 

3. The Federal Acquisition Regulation should be amended to 
describe specifically the full range of dispute resolution methods 
available for consideration by the parties at or before the time a 
claim is presented to the CO for resolution under the Contract 
Disputes Act. 



RECOMMENDATION 89-2, CONTRACTING OFFICERS' ADR 15 

4. COs involved in the disputes process should be specifically 
evaluated, as part of the annual performance evaluation cycle, 
on their effectiveness in managing contract disputes. 



5. In addition to those techniques set forth in 
Recommendation 87-11, agencies should be encouraged to use 
the following specific methods in CO-level disputes: 

(a) Employing factfinding to offer an advisory decision, 
or designating a CO who was not involved in the disputed 
issues, or a particular distinguished government official or 
other knowledgeable person, to make an advisory decision; 

(b) Employing minitrial or other processes to permit a 
structured presentation of facts and arguments to the CO 
or other government officer with authority to settle; 

(c) Agreeing in advance that disputes arising under a 
particular contract will be voluntarily submitted to an 
expert or panel for a nonbinding opinion as soon as a 
disagreement occurs; and 

(d) Encouraging agency COs to employ the services of 
mediators or other neutrals to enhance negotiations to 
settle contract disputes. 

6. Board of Contract Appeals Judges should take greater 
advantage of opportunities to suggest returning to the CO cases 
which evidently should be pursued more vigorously for 
settlement. 

7. ADR training programs, for both industry and government 
personnel, should be Integrated into existing management 
training programs, as follows: 

(a) Training should focus on the use of these techniques 
as tools to improve the contract formation and contract 
administration process, so as to abate conditions which 
later lead to disputes, and to expedite decisionmaking 
under the Contract Disputes Act. 

(b) Training should reflect the fact that negotiation is a 
key dispute resolution method, and that most COs would 
become more effective professionals by devoting increased 
training and attention to these methods. The Federal 
Acquisition Institute and other government entitles 
specializing in acquisition training should devote increased 
attention to listening and communications skills, use of 
"interest" and "principled" rather than "positional" 



16 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

bargaining, and systematic attention to negotiation 
techniques. The training should also enable a CO to engage 
In meaningful discussion with a contractor by first working 
as a "team builder" to develop a coherent intraagency 
position that takes into account the views and needs of 
attorneys, auditors, program managers, engineers and 
others within the agency. Consistent with best 

management practice and the Packard Commission Report 
for greater efficiency in procurement, 6 the training should 
encourage the CO, even without the assistance of a third- 
party neutral, to avert appeals by reducing the number of 
situations where disputes, encumbered by internal 
disagreements or incoherent positions, are passed on to 
boards of contract appeals. 

(c) Professional organizations concerned with the public 
contract disputes process, such as the American Bar 
Association, Federal Bar Association, and National Contract 
Mcinagement Association, should develop and encourage 
increased learning opportunities in effective dispute 
resolution techniques for representatives of the government 
and private sector. 



^A Quest for Elxcellence, Final Report by the President's Blue Ribbon 
Commission on Defense Management (June 1986). 



Recommendation 89-3 

Conflict-of-interest 

Requirements for Federal Advisory Committees 



The law and practice regarding conflict-of-interest 
requirements for federal advisory committee members have 
developed from the interaction of three statutory schemes: the 
Federal Advisory Committee Act.i the conflict-of-interest laws. 
and the federal personnel laws. However, none of these 
statutory schemes was drafted to deal specifically with conflict- 
of-interest standards for government advisers. 

In 1982 the Office of Government Ethics issued guidance to 
agencies that sought to meld a coherent analytical framework 
from the three statutory schemes. In determining whether the 
conflict-of-interest laws applied, the Office distinguished 
between those advisers who were selected as committee 
members because of their individual qualifications, and were 
thus deemed to be special government employees (SGE's), and 
those who instead were selected as representatives of 
nongovernmental groups or organizations (or in some cases, as 
independent contractors). While this guidance has reduced the 
confusion somewhat, the determination of a committee 
member's status as an SGE or a representative of a 
nongovernmental group or organization remains difficult, and 
agency practice in classifying advisory committee members as 
SGE's or representatives varies greatly and often appears 
arbitrary. 

The classification of an advisory committee member as an 
SGE or a representative is significant because only the former 
are subject to the conflict-of-interest and financial disclosure 
laws. The most significant of these laws for advisory committee 
members is Section 208 of Title 18, United States Code, which 
makes it a criminal offense to participate "personally and 
substantially" as a government employee "through decision, *** 
recommendation, the rendering of advice. Investigation, or 
otherwise in *** any particular matter In which to his 
knowledge, he, his spouse, minor child, partner, organization *** 

I5 U.S.C. App. I 



18 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

has a financial interest." The term "particular matter" in Section 
208 has been interpreted broadly by the Department of Justice 
and the Office of Government Ethics to extend to all discrete 
matters that are the subject of agency action, including 
rulemaking and general policy matters. 2 

Section 208 is especially a problem for advisory committee 
members. Often they have been selected precisely because they 
are especially well qualified to provide advice concerning 
problems in a particular field in which they themselves may be 
active both professionally and financially. 

Because of its breadth. Congress provided for agency waivers 
of section 208's prohibition, either by rule or on a case-by-case 
basis, where the appointing official makes a determination that 
the employee's interest is too remote or insubstantial to affect 
the integrity of his or her services. Agencies, however, may be 
unable or reluctant under current law to grant a waiver where a 
financial interest is significant, even though the agency 
concludes that any bias arising from that interest will be offset 
through committee balance, disclosure of the interest, or the 
individual's status as only an adviser and not as a 
decisionmaker. 

Faced with the specter of criminal liability and the limitations 
of waivers, or simply for administrative convenience, some 
agencies have adopted a policy of declaring most or all of their 
advisory committee members to be interest group 
representatives, rather than SGE's, except in the clearest cases. 
Thus, in practice, agencies may be requiring too little disclosure 
from members who are not SGE's, while imposing significant 
burdens, principally potential criminal liability, on those 
members who are SGE's. 

In this recommendation the Conference urges the 
establishment of a uniform minimal disclosure requirement for 
all advisory committee members, whether or not they are 
classified as SGE's. 3 The recommendation seeks to balance the 



2The test of whether a financial Interest exists with respect to the matter is 
whether the government action in which the employee participates will have a 
"direct and predictable effect" on the entity in question. Participation in the 
presence of a known conflict constitutes a violation of section 208, whether or 
not the employee's action furthers or is likely to further his or her financial 
interest. 

^The Conference recognizes that advisory committee members who are 
classified as special government employees may be required to furnish financial 
information pursuant to regulations of the appointing agency or the OfBce of 
Government Ethics. It is further noted that the Ofilce of Government Ethics has 



RECOMMENDATION 89-3, ADVISORY COMMITTEES & ETHICS LAWS 19 

government's and the public's need for information to evaluate 
potential conflicts of interest and the burden placed on the 
individual who agrees to serve on an advisory committee, 
frequently without pay. 

The Conference also recommends that Congress direct 
agencies to determine, when chartering or renewing the charter 
of an advisory committee, whether or not the committee's 
responsibilities require identifying its members as special 
government employees for purposes of the conflict-of-interest 
laws. The recommendation ( 2) includes criteria for making 
this determination. This approach places the burden of 
foreseeing and preventing conflicts of interest on the agency that 
seeks an individual's services on an advisory committee, rather 
than on the individual asked to serve, as does reliance on 208 
waivers. 

This recommendation does not extend to privately 
established advisory committees that are utilized for advice in 
particular matters because the members of these committees 
are not appointed by a federal agency. Consequently, an 
agency's relationship with such committees must be considered 
on a ad hoc basis. Nevertheless, the Conference believes 
agencies should be alert to possibilities for bias or self-interest 
in the advice of utilized committees cind, where appropriate, 
should request information respecting the affiliations and 
interests of the members. 



RECOMMENDATION 



1 . Disclosure by Advisory Committee Members. 

(a) Congress should require that each individual 
selected to serve on a federal advisory committee, excluding 
a regular government employee, furnish to the agency or 
appointing authority at the time of the appointment or 
designation — 

(1) the identity of the individual's principal 
employment; 

(2) a list of positions held (whether paid or unpaid) 
and any contractual relationships for the performance 



under consideration a proposed regulation governing financial disclosure for all 
government employees, including special government employees. 



20 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

of services with any corporation, company, firm, 
partnership or other business enterprise, any non- 
profit organization, any labor organization, or any 
educational or other institution whose activities or 
purposes may be (or may foreseeably become) relevant 
to the purposes and functions of the advisory 
committee as determined by the agency or appointing 
authority and described in the committee charter; 

(3) the identity, but not value or amount, of any 
other sources of Income or any interests in a trade or 
business, real estate, or other asset held for investment 
or production of income, exceeding $1,000 in value 
which are relevant to the purposes and functions of the 
advisory committee as determined by the agency or 
appointing authority and described in the committee 
charter; 

(b) Advisory committee members should be required to 
fQe updated disclosure reports annually. 

(c) The agency or appointing authority should make 
publicly available the Information furnished pursuant to 
subparagraphs (a)(1) and (a)(2) above. The financial 
information described in subparagraph (a)(3) should 
ordinarily be held confidential unless the member consents 
to its release or the agency determines after consulting with 
the member that public disclosure is required In the public 
interest. 

2. Classtfication of Advisory Committee Members. 

Congress, by amendment to the Federal Advisory 
Committee Act or other pertinent statute, should require 
that each agency determine, when chartering or renewing 
the charter of an advisory committee, whether its 
responsibilities are such as to require some or all of its 
members to be Identified as special government employees 
for purposes of the conflict-of-interest laws. Congress 
should require the agency to consult with the Office of 
Government Ethics in making such a determination, and it 
should direct the agency to be guided by the following 
considerations — 

(a) Ordinarily, where an advisory committee is expected 
to provide advice of a general nature from which no 
preference or advantage over others might be gained by a 
particular person or organization, the members of the 
committee need not be special government employees. 



RECOMMENDATION 89-3, ADVISORY COMMITTEES & ETHICS LAWS 21 

(b) The members of an advisory committee which 
renders advice with respect to the agency's disposition of 
particular matters involving a specific party or parties 
should be considered special government employees, 

(c) The principal consideration in classifying an 
advisory committee member should be the nature of the 
committee's function rather than whether or not the 
member receives compensation. 

3. Coverage. 

This recommendation applies to advisory committees 
which are established and whose members are appointed or 
designated by the federal government, and to advisory 
committees whose operations are funded by the 
government. It does not apply to privately established 
advisory committees which are "utilized" by the federal 
agencies in particular matters. 

4. Technical Amendment 

Congress should amend 18 U.S.C. 207 (g) to provide 
that a partner of a special government employee shall not 
be barred from any representational activity because of that 
employee's participation in a particular matter where the 
employee himself would not be barred from such 
representation by 18 U.S.C. 203 or 205. 



Recommendation 89-4 

Asylum Adjudication Procedures 



Providing asylum to the persecuted is a vital and treasured 
part of the American humanitarian tradition. It deserves 
reaffirmation and continued commitment. The asylum process, 
however, can also become a misused exception in the nation's 
immigration laws, especially in a time of improved 
transcontinental travel and communications. Two important 
public values thus come into conflict in the asylum program. 
On the one hand stands the promise of refuge to the persecuted; 
on the other stands the demand for reasonable assurance of 
national control over the entry of aliens. This tension becomes 
acute whenever application numbers rise. 

In the 1970s, the United States received approximately 2000 
applications for asylum each year. By 1988, that number had 
risen to approximately 60,000 applications. The Immigration 
and Naturalization Service (INS) projects 100,000 applications in 
1989. Government expenditures for coping with the increase 
have risen rapidly, both for adjudication and for detaining or 
otherwise arranging to shelter and feed the applicants. But this 
is necessarily only a stopgap measure. It would be far more cost 
effective in the long run to devote the resources necessary to 
improve asylum adjudication procedures. 

Although it should be possible to distinguish qualified from 
unqualified asylum applicants and thereby both honor the 
humanitarian tradition and avoid misuse of the asylum 
provision, several factors hinder our ability to do so. First, the 
"well-founded fear of persecution" standard, upon which asylum 
is based, is far from self-defining; there is no uniform 
understanding of its application to particular cases. Second, 
judgments about the relative risks faced by asylum seekers 
upon return to their native countries are unavoidably affected 
by preconceptions about what conditions may be like in those 
countries. It may also be misleading to posit a sharp distinction 
between economic migrants and political refugees. Asylum 
seekers represent a spectrum of motivations, and many leave 
their home countries because of a mix of political and economic 
reasons. Third, the facts upon which adjudication must rest are 
elusive, largely because they turn on conditions in distant 



RECOMMENDATION 89-4, ASYLUM ADJUDICATION 23 

countries. Moreover, the individual applicant, often inarticulate 
and uneasy, may be the only available witness to the specific 
events that underlie the claim. Therefore, credibility 
determinations can be crucial, but they are complicated by 
barriers to effective cross-cultural communication. 
Improvements in the system must make allowance for all these 
difficulties. 

The central standard for determining whether an applicant 
will be granted asylum derives from the definition of "refugee" 
contained in a United Nations (UN) treaty, the 1951 Convention 
relating to the Status of Refugees, amended by its 1967 
Protocol. Under section 208 of the Immigration and Nationality 
Act (INA), the Attorney General may, in his discretion, provide 
asylum to applicants who establish that they have a "well- 
founded fear of persecution" in the home country because of 
race, religion, nationality, membership in a particular social 
group, or political opinion. Additionally, section 243(h) of the 
INA establishes a mandatory country- specific protection which 
is known as nonrefoulement Section 243(h) provides that the 
government may not return an alien to a country where his "life 
or freedom would be threatened" on any of the same five 
grounds. Under current administrative practice, the most 
Important test has become the "well-founded fear" standard, 
because people granted asylum status are necessarily shielded 
against removal from the United States. 

Historically, the United States has employed a mix of 
adversarial and nonadversarial procedures for deciding on 
asylum and nonrefoulement claims. Currently, "walk-in" claims 
are adjudicated by examiners in the district offices of the INS 
after an essentially nonadversarial interview. It typically lasts 
about twenty minutes as the interviewer reviews the application 
form (1-589) and the applicant's supporting information, and 
also prepares and issues work authorlisation papers (provided 
that the claim is adjudged "nonfrlvolous"). The file is then sent 
to the State Department for its advisory views. The applicant is 
given fifteen days to respond to any recommendation by the 
State Department to deny the application. Subsequently, an 
INS examiner will review the file and issue a decision. This 
process may take eight months or more. Informal review of 
district office decisions is provided by the Asylum Policy and 
Review Unit (APRU), a small office in the Department of Justice 
created in April 1987. 

Denials in the district office are not appealable, but 
unsuccessful applicants may renew the application in 
adversarial exclusion or deportation proceedings before an 



24 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

immigration judge, who will consider the matter de novo. These 
judges are officials in the Executive Office of Immigration Review 
(EOIR), which is wholly separate from INS but is also a part of 
the Department of Justice. Aliens who do not fUe for asylum 
until such proceedings have started have no access to the 
district office; they will be heard only by an immigration judge. 

The immigration judge's ruling on asylum is appealable to 
the Board of Immigration Appeals (BIA), which is also located in 
EOIR. Appeals can easily consume a year or more, largely 
because of delays in receiving transcripts of immigration court 
hearings. No further administrative appeals are possible at the 
instance of the applicant, but on rare occasions, cases are 
considered by the Attorney General personally upon certification 
or referral. Judicial review of individual asylum denials almost 
always occurs as part of the review of exclusion or deportation 
orders under section 106 of the INA. 

Administrative adjudication alone involves five distinct 
administrative units (the District Office, the State Department, 
APRU, the Immigration Judges, and BIA), only two of which see 
the applicant in person. This multiplicity of agencies spreads 
resources thin, resources that should be concentrated efficiently 
so as to improve the quality of the procedure and assure that 
genuine refugees are granted asylum. 

Adjudication of an asylum claim through the various admini- 
strative and judicial levels requires several months and often 
consumes years. Such delays increase the attraction for 
marginal applicants because applicants Ccin enjoy substantial 
benefits, including work authorization and freedom of 
movement, throughout the period their claim is pending. 
Deterrents such as detention or limitations on work 
authorization could be used to minimize this magnet effect. 
Those measures, however, carry substantial disadvantages. 
Primarily, they are indiscriminate in their impact and may fall 
most heavily on genuine refugees who have already suffered 
greatly. These measures also entail higher costs for the federal 
government, especially when asylum claims remain pending for 
lengthy periods. 

The Conference believes that fair but speedy conclusion of 
adjudication, leading either to a grant of asylum or to an 
enforceable removal order, is crucial to any healthy asylum 
adjudication system. This objective can be promoted through 
attention to two elements. First, delay derives in part from the 
provision of two separate rounds of de novo consideration of 
asylum claims. One unified Initial asylum proceeding should be 
established instead. (If the alien has other defenses to 



RECOMMENDATION 89-4, ASYLYM ADJUDICATION 25 

deportation or exclusion, those other defenses should continue 
to be heard by immigration Judges in contemporaneous and 
separate proceedings). Second, additional delay derives from 
the qualified right to counsel as specified by current statutes 
and regulations, which provide for counsel in exclusion or 
deportation cases "at no expense to the government." Because 
so many applicants are indigent, delays often result from the 
need to accommodate the schedules of those attorneys who are 
willing to take the cases on a pro bono basis — a problem that 
is compounded when applications increase in a particular 
geographic location. A healthy system of asylum adjudication 
must be able to schedule hearings expeditiously, even if pro 
bono counsel are not immediately available in sufficient 
numbers. Fairness must be sought, therefore, through hearing 
procedures, training, and monitoring that assure a special role 
for the adjudicator in developing a complete record when the 
applicant is not represented. 

The Conference also believes that a healthy asylum 
adjudication process must foster the greatest possible accuracy 
as well as public confidence that decisions are rigorous, 
professional, and unbiased. Reliance on a specialized 
adjudicative board without routine reference of applications to 
the State Department would serve these ends and minimize any 
perception that asylum decisions are influenced by political 
considerations. Additionally, arrangements must be made to 
provide the adjudicators with information concerning foreign 
country conditions that is as accurate and complete as possible, 
derived from a wide variety of sources, both to help dislodge any 
preconceptions and to foster systematic expertise for use in 
developing the record and making the ultimate judgment on the 
claim. 

For several years the Department of Justice has been 
considering amended asylum regulations that would serve many 
of these ends. Aversion proposed in August 1987 [52 Fed. Reg. 
32552] would have established a specially-trained corps of 
adjudicators, responsible to the INS Central Office rather than 
to the district directors, and it would have eliminated de novo 
reconsideration of asylum claims by immigration judges. These 
regulations drew criticism, in part because of concern about the 
professionalism and independence of the adjudicators, and the 
Department responded with modified proposed regulations in 
April 1988 [53 Fed. Reg. 11300] that retained the new corps of 
adjudicators but also restored the availability of de novo 
consideration before the immigration judges. Those regulations 



26 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

are still pending in the Attorney General's office and the 
Department has encouraged this study and analysis. 



RECOMMENDATION 

The Attorney General should adopt regulations creating a 
new asylum adjudication process that would eliminate much of 
the duplication and division of responsibility associated with the 
current complicated system. Resources should be applied to 
enhance the professionalism, independence, and expertise of the 
adjudicators, and to assure fair and expeditious adjudications, 
so that genuine refugees may be speedily given a secure status 
and unqualified applicants, absent circumstances which would 
allow them to remain in this country, may be promptly 
deported. 

I. Creation of a New Asylum Board 

The Attorney General should create a new Asylum Board 
located, for administrative purposes, within the Executive 
Office of Immigration Review (EOIR) of the Department of 
Justice and consisting of an adjudication division, an 
appellate division, and a documentation center. The 
chairperson of the Asylum Board would be responsible for 
administrative support and supervision of the operations of 
all three units. 

A. The Adjudication Division 

1. Jurisdiction. All claims for asylum under 
section 208 of the Immigration and Nationality Act 
(INA) or withholding of deportation under INA section 
243(h) (hereinafter collectively "asylum" claims) should 
be heard exclusively by asylum adjudicators in the 
adjudication division of the Asylum Board. 

2. Nature of the asylum hearing. Asylum claim 
proceedings should be recorded.! The asylum 
adjudicator should be responsible for developing a 
complete record of the specific facts relating to the 
applicant's claim, including those which might support 
a grant of asylum and those which might cast doubt on 



^The Administrative Conference recommends experimentation with other 
methods for creating a record that would maintain flexibility but preserve 
objectivity, professionalism, and fairness to the applicant. 



RECOMMENDATION 89-4, ASYLUM ADJUDICATION 27 

the claim or on the applicant's credibility. Care should 
be taken to assure the service of skilled Interpreters. 
The adjudicator should be responsible for most of the 
questioning, with a reasonable and adequate 
opportunity for additional questioning and entry of 
relevant information, including the presentation of 
witnesses, by the applicant and counsel. The 
Immigration and Naturalization Service (INS) should 
not be represented as an opposing party in the 
proceedings. 2 

3. Representation of applicants. Applicants 
should be encouraged to secure counsel (or a qualified 
nonattomey representative) to develop the initial claim 
and to provide representation during the asylum 
proceeding. Although reasonable accommodation 
should be provided for counsel to be obtained, 
proceedings should not be unduly delayed, because 
expeditious initial decisions are essential. 

4. Use of offlcial notice of country conditions. 
Asylum adjudicators should develop substantial 
cumulative expertise regarding country conditions, to 
be used in developing the record, and should be 
responsible for posing illuminating questions to the 
applicant and other witnesses, for evaluating evidence, 
and for reaching the ultimate determination about 
likely risks to the applicant upon return to the home 
country. The accepted standards for official notice, in 
accordance with the Administrative Procedure Act, 
should govern use of such Information. Ordinarily, 
these standards will simply require an adequate 
statement of reasons for accepting or rejecting the 
asylum claim, reflecting such expertise. In instances 
when specific and detailed facts developed from the 
documentation center or other sources (and not from 
information supplied by the applicant) appear to be 
crucial, the applicant should be given notice of intent to 
deny based on such Information, along with an 
opportunity to offer information or argument in 
rebuttal. 

5. The adjudicators. Asylum adjudicators should 
be recruited from among attorneys possessing 
adjudicative skills and appropriate judgment and 



^The Administrative Conference takes no position on the possible application 
of the Equal Access to Justice Act to asylum proceedings. 



28 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



temperament, with close attention given to those who 
are familiar with international relations and refugee 
affairs and who are sensitive to the difficulties of cross- 
cultural communication. Adjudicators should receive 
salary, benefits, and guarantees of adjudicative 
independence equivalent to those of immigration 
judges, and they should be assigned no other 
enforcement or adjudication responsibilities. The 
adjudicators should be given thorough and ongoing 
training, especially on techniques for fairly conducting 
this specialized type of proceeding and on conditions in 
those countries from which a substantial number of 
asylum applications is received. If, alternatively, a 
separate Asylum Board is not created and the 
adjudication assignment is given to immigration 
judges, then such judges should be assigned to a 
separate unit in EOIR. 
B. Appellate Division 

1. Composition and functions. The appellate 
division of the Asylum Bocird should consist of the 
chaiirperson and two additional members, assisted by 
staff attorneys and other support personnel. The 
division's principal responsibilities should be to 
consider appeals filed by persons denied asylum at the 
initial stage, in Ught of the administrative record 
compiled before an adjudicator, and such other 
information as the applicant may wish to submit or of 
which official notice may be taken. The division, 
however, should also monitor cases, and should have 
the authority to require certification to it of selected 
cases, either granting or denying asylum, in order to 
foster consistency, fairness, and political neutrality. It 
will thus absorb the principal functions now performed 
by the Asylum Policy and Review Unit. 

2. Certification or referral to the Attorney 
General. The Attorney General should retain the 
authority to review decisions of the Asylum Board, 
upon formal certification or referral or sua sponte. 

3. Expeditious completion of appeals. A high 
priority should be placed on completing all asylum 
appeals expeditiously, preferably within three months 
of filing. The Department of Justice should ensure that 
transcripts, where required, are made from recorded 
hearings in a timely fashion. 



RECOMMENDATION 89-4, ASYLYM ADJUDICATION 29 

C. Documentation Center 

A documentation center, staffed with regional 
specialists, should maintain current and detailed 
information on country conditions, from both 
governmental and nongovernmental sources, 
periodically compile and publish usable summaries on 
selected countries, and respond to requests for more 
specific information received from officials of the 
Asylum Board. Special effort should be devoted to 
assuring complete compilations of ongoing reports from 
established nongovernmental human rights 
organizations, and to drawing upon information from 
documentation centers in other countries. Information 
and procedures developed by other countries can be 
particularly useful in minimizing start-up costs. The 
center's collections and publications shall be accessible 
to the public. 

D. Role of the Department of State and the United 
Nations High Commissioner for Refugees 

The Department of Justice should take advantage of 
resources, assistance, and information available 
through the State Department and the United Nations 
High Commissioner for Refugees (UNHCR). In 
particular, arrangements should be made with both to 
assist in training adjudicators and to augment 
information available through the documentation 
center. 

ff it so requests, on an across-the-board or country- 
specific basis, the State Department should receive 
notice of individual asylum applications, so that it may 
offer its Judgment, in particular, about appropriate 
responses in sensitive cases, such as those involving 
foreign government officials. 

II. Detention 

Where detention of asylum seekers is deemed 
necessary,3 the Department should limit it to short-term 
detention in "asylum processing centers," as recommended 
by the Select Commission on Immigration and Refugee 
Policy. Such centers should also keep families together 
wherever possible, minimize the length of detention, provide 



»^The Administrative Conference does not take a position on the suitability of 
detention in asylum proceedings. 



30 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

assistance in securing representation, and otherwise foster 
conditions which reflect that the purpose of detention is not 
punitive. 

III. Deportation 

The Department of Justice should ensure that 
individuals denied asylum are removed promptly if they are 
otherwise excludable or deportable, subject to any policy 
decision by the Attorney General to grant extended 
voluntary departure to nationals of particular countries. 

IV. Judicial Review 

Judicial review of asylum denials should be available as 
part of the review under section 106 of the INA for orders of 
deportation or exclusion. Appropriate arrangements 
therefore should be made to combine, for purposes of 
judicial review, the record of proceedings before the Asylum 
Board with that of the regular deportation or exclusion 
proceedings before the Immigration Judges and the Board of 
Immigration Appeals. 



Recommendation 89-5 

Achieving Judicial Acceptance of Agency 
Statutory Interpretations 



Agencies continually interpret the statutes they administer. 
Their interpretations are expressed in a great variety of formats - 
- including, among others, legislative regulations, adjudicatory 
opinions, court briefs, interpretive rules, policy statements, staff 
instructions, correspondence, informal advice, press releases, 
guidance manuals, testimony before Congress, speeches, and 
internal memoranda. This recommendation addresses the 
relationship between the procedures used by an agency in 
interpreting a statute and the role of the courts in statutory 
interpretation. 

Interpretation of a statute presents a question of law, 
traditionally the province of the judicial branch (see the scope of 
review provision of the APA, 5 U.S.C. 706). However, for many 
years courts have accorded respectful attention or even 
controlling effect to interpretations of statutes made by the 
agencies that administer them. In some situations. In which the 
courts reserve the power to arrive independently at their own 
interpretations, they will give respectful consideration to an 
agency's construction but may reject it, even if it seems 
reasonable. In other cases, courts consider themselves bound 
to accept an agency's interpretation outright, provided only that 
it is consistent with the statute and is reasonable. The law 
governing judicial acceptance of agency statutory interpretations 
is now dominated by Chevron U.SA. v. Natural Resources 
Defense Council 467 U.S. 837 (1984). In that case, one 
involving legislative rulemaking, the Supreme Court laid out a 
general framework for reviewing agency interpretations of 
statutes. First, the court is to determine whether Congress has 
directly spoken to the precise question at issue. If the intent of 
Congress is clear, the court (like the agency) must give effect to 
the congressional intent. Where Congress' intent is not clear, 
however, the court must determine whether the agency's 
interpretation is based on a reasonable construction of the 
statute. Chevron thus requires a reviewing court to accept an 
agency interpretation that (a) is not contrary to statute or 
specific statutory intent and (b) Is reasonable. 



32 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

When an agency issues a legislative rule or interprets its 
statute in a formal adjudication, its interpretation of the statute 
it administers is entitled to judicial acceptance under the 
Chevron standard. Similarly, acceptance under the Chevron 
standard is appropriate if the reviewing court finds a 
congressional delegation of authority to make definitive 
interpretations in an informal format such as the informal 
agency staff ruling involved in Ford Motor Credit Co. v. MUhoUin, 
444 U.S. 555 (1980). But agencies rarely possess 
congressionally delegated authority to make definitive 
interpretations, carrying the force of law, by informal means. 
Thus, when an agency states its interpretation of a statute in an 
informal format, it should understand that courts ordinarily will 
not be bound to accept such an interpretation. 

This is not to say that reviewing courts may ignore an agency 
interpretation set forth in an informal format. Numerous 
decisions of courts at all levels indicate that the views of the 
agencies charged with responsibility for administering a statute 
are accorded weight and may be highly influential in shaping 
courts' decisions. In this way courts retain the advantage of 
administrative agencies' expertise and remain free to adopt 
agencies' interpretations, even though not required to do so. 

Even when interpretations are expressed informally, however, 
agencies have in some instances successfully asserted that 
these interpretations should be accepted as definitive by the 
courts, without consideration of whether the agency possesses 
the authority to make binding interpretations in the format it 
has used. 

When an agency interprets a statute without using 
procedures authorized by Congress for the development of 
definitive statutory interpretations, it should not expect that its 
interpretation will be entitled to judicial acceptance as definitive. 
Procedures so authorized by Congress, in almost all cases, will 
be relatively formal ones that ensure some level of public 
participation and encourage reasoned and thoughtful 
decisionmaking by the agencies. However, this recommendation 
is not intended to discourage agencies and their staffs from 
using informal means to keep the public apprised of their views 
on questions of statutory interpretation. It is often useful and 
appropriate for agencies to provide informal guidance of this 
type. The agency may reasonably expect that interpretations 
like these are entitled to such special consideration as their 
nature and the circumstances of their adoption warrant. But It 
Is Important for both agencies and courts to remember that 



RECOMMENDATION 89-5, STATUTORY INTERPRETATIONS 33 

these informal expressions should not be accorded the same 
weight as definitive agency interpretations. 

This recommendation relates solely to the procedures that 
should be preconditions to agencies* assertion of the Chevron 
standard of review. It thus takes no position concerning any 
other aspect of the Chevron standard. 

Accordingly, the Administrative Conference recommends that 
the following process be observed. 



RECOMMENDATION 



In developing an interpretation of a statute that is intended to 
be definitive, an agency should use procedures such as 
rulemaking, formal adjudication, or other procedures authorized 
by Congress for, and otherwise appropriate to, the development 
of definitive agency statutory interpretations. 



Recommendation 89-6 

Public Financial Disclosure by Executive Branch 
Officials 



Public financial disclosure by federal officials Is Intended to 
make It possible to monitor actual or potential conflicts of 
Interest of such officials. This. In turn, may deter public 
officials from even considering conduct that would present the 
appearance of a conflict of Interest. However, these benefits of 
public financial disclosure must be balanced against the 
burdens Imposed on the federal officials who are subject to 
them. 

Determining appropriate public financial disclosure 
requirements requires an assessment and accommodation of 
three concerns: the relevance of the Information to conflicts of 
Interest which might be faced by the individual in his or her 
official capacity: the practical burden faced by an individual who 
must assemble and report information accurately (including 
whether a nominee or employee would reasonably be expected 
to have at hand the information which he or she is required to 
report): and the psychological burden imposed on an Individual 
who must make his or her financial status publicly available to 
others (i.e., whether public disclosure constitutes an excessive 
invasion of privacy). 

The Administrative Conference has studied the Ethics in 
Government Act's executive branch financial disclosure 
requirements (codified at 5 U.S.C. 201-209) and in this 
recommendation urges Congress to make specific changes to 
those requirements, consistent with an appropriate balance of 
the benefits and costs of such disclosure. 

This recommendation is not made with the intention of 
generally requiring either more or less disclosure of public 
officials. Rather, the Conference's goal is to rationalize the 
Ethics in Government Act's requirements and eliminate those 
that appear to bear no reasonable relationship to the Act's 
purposes. On the one hand, the recommendation Increases 
disclosure by reducing the current threshold level for the 
reporting of a covered individual's liabilities from $10,000 to 
$1,000, to be consistent with the current threshold level of 
$1,000 for the reporting of assets ( 2 b (1)). On the other hand. 



RECOMMENDATION 89-6, EXECUTIVE BRANCH FINANCIAL DISCLOSURES 35 



the recommendation would lessen disclosure by reducing the 
number of categories of value for the reporting of assets from 
the current six to two, which the Conference believes is 
sufficient for conflict-of-interest analysis and the maintenance of 
public confidence in the integrity of executive branch officials ( 
2 b (2)). 

Because the Act's executive branch financial disclosure 
provisions are so detailed, this recommendation has been 
organized to clearly distinguish between current provisions that 
the Conference believes generally further the Act's purposes 
and, therefore, should be retained, and those provisions that 
appear unnecessary to achieve the Act's purposes and, 
therefore, should be eliminated or changed. However, in 
recommending the retention of particular provisions, the 
Conference does not mean to imply that such provisions cannot 
be improved. To the contrary, the Conference urges the 
Congress to systematically review the coverage and language of 
all of the Act's public financial disclosure provisions, and to 
rewrite those that can be made clearer and simpler. 

To Illustrate, the Conference recommends continuation of the 
current requirement that nominees for positions covered by the 
Act report the source of all earned income in excess of $5,000 
received by a reporting individual from one source in the two 
years preceding the year of filing ( 2 a (2)). However, the 
current statutory provision (5 U.S.C. 202(a)(6)(B)) requires 
reporting of such compensation paid "in any of the two calendar 
years prior to the calendar year during which the individual files 
his first report . . . ." If strictly applied, a nominee who filed a 
report in October of 1989 would be required to disclose such 
compensation for calendar years 1987 and 1988, but not for the 
period in 1989 prior to his or her entering government service. 
This theoretical gap in coverage should be closed whether or not 
in practice it has proven to be a problem. 

The same statutory provision exempts from the "over-$5,000 
from one source" disclosure requirement the reporting of "any 
information with respect to any person for whom services were 
provided by any firm or association of which such individual 
was a member, partner, or employee unless such individual was 
directly involved in the provision of such services." 5 U.S.C. 
202 (a)(6)(B) (emphasis added). In redrafting this provision. 
Congress should consider either defining the term "directly 
involved" or delegating to the Office of Government Ethics the 
responsibility to clarify its meaning by regulation, especially as 
applied to individuals who provide services to others, such as 
lawyers. Therefore, although the Conference supports the 



36 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



retention of the substance of this and other of the Act's financial 
reporting provisions, it is clear that improvements to the 
language and coverage can be made. 

Because of its limited mandate, the Conference takes no 
position on the public financial disclosure requirements 
applicable to legislative and judicial branch officials. However, 
the similarity of those requirements to executive branch 
requirements suggests the desirability of reviewing and possibly 
amending legislative and judicial branch requirements as well. 



RECOMMENDATION 



Persons Required To File. 

a. Positions For Which Coverage Should Be Retained. 
Congress should continue to require the following 
categories of executive branch personnel to make public 
financial disclosure: 

(1) The President, Vice President, and nominees for 
and incumbents in positions which require Senate 
confirmation; 

(2) Full-time officers and employees of the executive 
branch (including independent agencies) whose 
positions are classified as OS- 16 or above or who are 
paid at or above the minimum rate of pay fixed for GS- 
16; 

(3) Each member of a uniformed service whose pay 
grade is at or in excess of 0-7; 

(4) The Postmaster General, Deputy Postmaster 
General, each Governor of the United States Postal 
Service, and each Postal Service and Postal Rate 
Commission officer or employee whose rate of pay 
equals or exceeds the minimum rate of basic pay for 
GS-16; 

(5) Each administrative law judge appointed 
pursuant to 5 U.S.C. 3105; and 

(6) All other employees determined by the Director 
of the Office of Government Ethics to be in positions 



^The Conference is authorized by statute to study and make 
recommendations relating to administrative procedure used by administrative 
agencies in carrying out administrative programs, 5 U.S.C. 574. 



RECOMMENDATION 89-6, EXECUTIVE BRANCH FINANCIAL DISCLOSURES 37 



equal in responsibility to those normally classified at 
GS-16 or above. 

b. Positions For Which Coverage Should Be Removed. 
Congress should amend the Ethics in Government Act to 
remove the reporting requirement, except as may be 
required under subsection c below, from the following 
persons: 

(1) Candidates for the offices of President and Vice 
President who are not receiving federal funds under the 
federal election laws and who are not government 
officials otherwise required to report;2 

(2) Special government employees; 3 and 

(3) Designated agency ethics officers whose rate of 
pay or other responsibilities would not otherwise 
subject them to the reporting requirement. 

c. Administrative Extensions of Coverage. Congress 
should amend the Ethics in Government Act to permit the 
Director of the Office of Government Ethics to extend the 
reporting requirement, on a position or categorical basis, to 
any officer, employee or special government employee of the 
executive branch not covered by the Act, whose position is 
determined by the Director to present an unusual 
opportunity for conflicts of interest. 

d. Administrative Exemption From Coverage. Congress 
should amend the Ethics in Government Act to permit the 
Director of the Office of Government Ethics to exempt from 
the reporting requirement those positions included in 
subsection a above whose responsibilities are identified by 
their agencies and determined by the Director to be unlikely 
to place their incumbents in situations of conflict of 
interest. 

e. Review of Coverage Extensions and Exemptions. 
Congress should require the Office of Government Ethics 
annually to review, based on the recommendation of the 
designated agency ethics officials, all determinations 
currently in effect under c and d above. 



21116 Conference recognizes that candidates for these offices are not 
executive brsinch officials; nonetheless, this recommendation addresses coverage 
of candidates because they are included In the current statute setting forth 
executive personnel financial disclosure requirements. 

3lt is noted that the Administrative Conference has recommended minimal 
financial disclosure for all members of federal advisory committees, including 
those members who are special government employees. Sec ACUS 
Recommendation 89-3, Confllct-of-Interest Requirements for Federal Advisory 
Committees, 1 CFR 305.89-3. 



38 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



2. lNFX)RMATION REQUIRED TO BE FILED. 

a. Reporting Requirements That Should Be Retained. 
Congress should leave the Ethics in Government Act 
unchanged in the following respects: 

(1) Reporting by Both Incumbent and Nominated 
Officials. Congress should continue to require both 
incumbent executive branch officers and employees 
whose positions are covered by the Ethics in 
Government Act, and nominees for those positions, to 
disclose publicly the following categories of information: 

(a) the identity of any interest in a trade or 
business or asset held for investment or 
production of income, if the vsdue of the interest 
exceeds $1,000; 

(b) the identity of all positions held by the 
reporting individual as an officer, director, 
trustee, partner, proprietor, representative, 
employee or consultant of any corporation, 
company, firm, partnership, or other business 
enterprise, any non-profit organization, any labor 
organization, or any educational or other 
institution other than the United States, but not 
Including positions held in religious, social, 
fraternal, or political entities, or positions solely 
of an honorary nature; and 

(c) the date, parties to, and terms of any 
future employment arrangements negotiated by 
the reporting individual, leaves of absence during 
the period of federal service, continuing payments 
from a former employer, or continuing 
participation in a former employer's welfare or 
benefit plan. 

(2) Reporting Only by Nominated Officials. In 
addition to the information required to be reported by 
incumbent and nominated executive branch officers 
and employees under subsection (1) above. Congress 
should continue to require that nominees for positions 
covered by the Ethics in Government Act report the 
source of all earned income in excess of $5,000 
received by the reporting individual from one source in 
the two years preceding the one in which the nominee 
files, and a brief description of the services for which 
the compensation was paid. As current law provides, 
this requirement should not apply to information about 



RECOMMENDATION 89-6, EXECUTIVE BRANCH FINANCIAL DISCLOSURES 39 

any person for whom services were provided by the firm 
or association of which the nominee was a member, 
partner, or employee, unless the nominee was directly 
involved tn the provision of such services. 

(3) Reporting Only by Incumbent Officials. In 
addition to the information required to be reported by 
incumbent and nominated executive branch officers 
and employees under subsection (1) above. Congress 
should continue to require covered incumbent 
executive branch officers and employees to disclose the 
following categories of information: 4 

(a) the source, type and amount of non- 
governmental earned income received by the 
reporting individual, including honoraria, which 
in the aggregate exceeded $100; and 

(b) the date and a brief description of each 
purchase, sale or exchange of real property, 
stocks, bonds, commodities futures or other 
property with a value over $1,000, except (i) 
transactions between the reporting individual and 
a spouse or dependent children, (ii) transactions 
involving a personal residence of the reporting 
individual or the individual's spouse, and (ili) 
transactions involving an investment in the 
nature of a cash equivalent [e.g., a money market 
fund, certificate of deposit, or personal bank 
account.) 

(4) Interests of Spouses and Dependent Children. 
The present statutory provisions on reporting of the 
interests of spouses and dependent children of the 
reporting official should be retained. 

b. Reporting Requirements That Should be Changed. 
Congress should amend the Ethics in Government Act to 
change the reporting requirements in the following ways: 

(1) Liabilities. The present requirement of reporting 
the identity of liabilities in excess of $10,000 owed by 
the reporting individual should be changed to a 
requirement of reporting liabilities in excess of $1,000, 
the same value which the statute now uses for 
reporting of assets. As present law provides, the 
reporting requirement should not extend to the 



^Under cvirrent practice individuals who joined the government in the 
preceding calendar year are only required to report this information for their 
period of government service and not before. 



40 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



individuars home mortgage, loans for the purchase of 
personal property which are secured by the property 
purchased and which do not exceed the value of the 
security, sums owed to a relative, and revolving charge 
accounts with a balance less than a specified amount 
at the end of the reporting period (currently $10,000). 

(2) Categories of Value. The present requirement 
that assets, liabilities, and transactions in assets above 
the $1,000 threshold be reported in numerous 
categories of value should be eliminated. However, in 
order to distinguish large interests from those of lesser 
significance, the reporting individual should be 
required to state whether each particular asset, liability 
or transaction was in excess of a specified higher 
amount [e.g., $50,000 or $100,000 each). 

(3) Sources of Earned Income Prior to Government 
Service. The requirement that all nominees for covered 
positions report the source, type and amount of non- 
government earned income which they received in the 
year prior to entering government service should be 
eliminated, except for amounts in excess of $5,000 
received from one source {see 2 a (2) above). 

(4) Income from Assets Otherwise Reported. The 
requirement that both incumbents and nominated 
officials report income in excess of $100 from each of 
their investments should be eliminated because the 
assets themselves are already reported. 

(5) Reimbursements and Gifts. 

(i) Reporting Period. The date after which all 
covered reimbursements and gifts should be 
required to be reported should be the date on 
which the official is nominated for or appointed to 
the position covered by the Ethics in Government 
Act, not the date the official takes office. 

(11) Reimbursement and Gifts of Travel or 
B^itertainment. The threshold amount for 
reporting reimbursements and gifts of 
transportation, lodging, food or entertainment, 
other than personal hospitality from an 
individual, received by the reporting individual 
from any source other than a relative during the 
reporting period should be changed from $250 
per year to a per event amount {e.g., $100 or 
$150) to avoid reporting de minimis information. 
The statute should be amended further to 



RECOMMENDATION 89-6, EXECUTIVE BRANCH FINANCIAL DISCLOSURES 41 

require, in addition to the source and a brief 
description, the reporting of the value or amount 
of such reimbursements or gifts in broad 
categories {e.g., under $1,000; $1,000 to 
$10,000; over $10,000) in accordance with 
regulations issued by the Office of Government 
Ethics. 

(iii) All Other Covered Gifts. The requirement 
of reporting all gifts to the reporting Individual, 
other than gifts of transportation, lodging, food or 
entertainment, which aggregated more than $100 
in value over the reporting period, excluding gifts 
from relatives of the reporting individual, and not 
aggregating gifts of $35 or less in calculating the 
$100, should be retained. However, the statute 
should be amended to require, in addition to the 
source and a brief description, the reporting of 
the value or amount of such gifts in broad 
categories [e.g., under $1,000; $1,000 to 
$10,000; over $10,000) in accordance with 
regulations issued by the Office of Government 
Ethics. 



Recommendation 89-7 

Federal Regulation of Biotechnology 



New biotechnology techniques promise great benefits in fields 
such as medicine, agriculture, and manufacturing. However, 
these new techniques, which involve alteration of the genetic 
structure of an organism, have raised concerns that some new 
organisms or products may be dangerous to individuals or 
detrimental to the environment. This recommendation 
addresses coordination of federal regulation in this area and the 
procedures agencies use to regulate biotechnology development, 
testing and use. 

Genetically-engineered organisms are regulated under a 
variety of statutes enacted to prevent or reduce society's 
exposure to unsafe or harmful products or substances. The 
agencies with such statutory authorities currently share 
responsibilities for regulation of biotechnology in accordance 
with policy statements issued in 1986 by the agencies and by 
the Director of the Office of Science and Technology Policy (who 
serves as the President's Science Adviser), i In its policy 
statement, the Office of Science and Technology Policy (OSTP) 
attempted to clarify the responsibility of each agency where 
more than one agency shared jurisdiction to regulate 
biotechnology areas. 

The Conference recommends a continuation of interagency 
coordination under the auspices of the OSTP. Experience does 
not currently indicate that new legislation is needed for effective 
interagency coordination of biotechnology regulation. On the 
other hand, the Conference believes that both the President and 
the Congress should monitor closely the coordination process 
because of the importance of this area to the nation's economic 
and social well-being. 

The Conference also recommends that the President and the 
Congress, through the OSTP and the Office of Technology 
Assessment (OTA), survey biotechnology developments and 
agency regulation of biotechnology under existing statutes to 
determine whether current law and regulation provide adequate 



lOSTTP, Coordinated Framework for Regulation of Biotechnology: 
Announcement of Policy and Notice for Public Comment. 51 F.R 23302 (1986). 



RECOMMENDATION 89-7, BIOTECHNOLOGY REGULATION 43 

authority to protect public and private interests or whether in 
particular instances current regulation is unnecessary. The 
survey should identify whether nonregulation of any particular 
area reflects an agency decision not to use its authority to 
regulate or an absence of regulatory authority. 

The Conference also urges changes in the coordination role 
for biotechnology regulation performed by the President's Office 
of Science and Technology Policy. Most importantly, the 
Conference urges the President to make the work of the Office's 
Biotechnology Science Coordinating Committee (BSCC) a high 
priority. A revitalized BSCC can help agencies coordinate their 
activities concerning biotechnology development, regulation, 
funding, and biosafety research. To fulfill this mandate, the 
Conference believes the BSCC's role should emphasize fact- 
finding, reporting, and serving as a clearinghouse for 
information relating to biotechnology. 

The Conference recommends that agencies engaged in 
biotechnology regulation articulate their policies through generic 
rules and policy statements to the extent possible. Since public 
acceptance of agency decisions is especially important In this 
area and because of the novelty and uncertainty of the risks 
associated with biotechnology, the Conference encourages 
agencies to adopt appropriate procedures to allow public 
participation. Agencies are also encouraged to seek ways to 
make biosafety information available to the public to the 
maximum extent consistent with protection of the proprietary 
interests of submitters of confidential business information. 



RECOMMENDATION 



Biotechnology Regulatory Structure 

(a) Interagency coordination is critically needed to 
mitigate problems caused by concurrent regulation of 
biotechnology by two or more agencies. The Office of 
Science and Technology Policy's Biotechnology Science 
Coordinating Committee (BSCC) should have primary 
responsibility for identifying issues, exchanging information 
and preparing reports concerning issues common to several 
agencies. Responsibility for establishing uniform 

government policies should be retained by the Office of 



44 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



Management and Budget working in coordination with the 
BSCC. 

(b) The President and Congress should survey 
biotechnology developments and agency regulation of 
biotechnology under existing statutes to consider whether 
and in what respects current regulation of biotechnology is 
inadequate or excessive. To facilitate this, the President's 
Office of Science and Technology Policy (OSTP) and 
Congress' Office of Technology Assessment (OTA) should, 
jointly or separately, identify all areas of biotechnology 
activity and determine the extent to which they are being 
regulated. OSTP and OTA should assess whether or not 
additional or diminished regulatory authority is desirable in 
such areas and furnish their findings and recommendations 
to the President and Congress. 

2. Regulatory Coordination 

(a) The President should make coordination of the 
government's activities relating to biotechnology a high 
priority. This should include: 

(1) monitoring the effectiveness of interagency 
coordination; 

(2) directing the Science Adviser to enlarge the 
membership of the BSCC to include all federal agencies 
that have substantial responsibilities for biotechnology 
research, development, or regulatory policy: and 

(3) directing the Science Adviser to invite 
representatives of other agencies to participate in the 
BSCC's activities, as appropriate, such as when their 
regulatory or other official responsibilities may be 
affected. 

(b) The BSCC should have a broad subject-matter 
mandate, including issues of biotechnology development, 
regulation, funding, and biosafety research. The 
Committee's role should emphasize fact-finding, reporting, 
and serving as a clearinghouse for information relating to 
biotechnology. 

(c) The Science Adviser should establish a policy for the 
BSCC that will foster opening its proceedings to the public. 

(1) Meetings of the BSCC should be open to the 
public unless they involve confidential information. 

(2) Members of the public should be allowed to 
provide comments to the BSCC either orally or in 
writing. 



RECOMMENDATION 89-7, BIOTECHNOLOGY REGULATION 45 

(3) The BSCC may invite advice from experts 
outside the government. 

(4) The BSCC should keep minutes or other records 
of its proceedings, including the reasons for closing any 
meetings. 

3. Regulatory Procedures 

(a) Agencies should, where appropriate, seek 
opportunities to promulgate generic biotechnology rules to 
address recurring regulatory issues. 

(b) Agencies should consider the adoption of rules or 
policy statements to enunciate the principles or criteria they 
will include in their risk assessment and management 
decisions. When adopting policy statements, agencies 
should follow the public participation procedures set forth 
tn Conference Recommendation 76-5.2 

(c) Agencies should consider adopting appropriate 
procedures to allow public participation and other forms of 
input when making regulatory determinations concerning 
biotechnology. Such procedures might include: 

(1) giving notice to the public with an invitation to 
submit comments concerning the determination; 

(2) providing additional notice of pending 
regulatory actions to persons who live near sites where 
proposed activities would take place; 

(3) holding informal public hearings to supplement 
written procedures; or 

(4) utilizing advisory committees under the Federal 
Advisory Committee Act. 3 

(d) Agencies should seek ways to meet the public's need 
for biosafety information about substances or organisms 
produced through biotechnology, without divulging 
confidential business information. 4 Such steps might 
include: 

(1) requesting submitters of confidential business 
information to focus their claims for confidentiality as 
much as possible; 



2aCUS Recommendation 76-5, Interpretive Rules of General Applicability 
and Statements of General Policy. 1 C.F.R 305.76-5 (1989). 

3See ACUS Recommendation. 82-5, Federal Regulation of Cancer-Causing 
Chemicals, Part IV. 1 C.F.R 305.82-5 (1989). 

4See ACUS Recommendation 82-1. Exemption (b)(4) of the Freedom of 
Information Act. 1 C.F.R 305.82-1 (1988). 



46 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



(2) requiring submitters of data that include 
confidential business information to identify those 
portions that are claimed to be confidential and to 
substantiate their claims at the time of submission; 
and 

(3) summarizing or aggregating confidential data in 
a manner that does not compromise confidentiality. 



Recommendation 89-8 

Agency Practices and Procedures for the Indexing 
and Public Availability of Adjudicatory Decisions 



This recommendation examines the obligation of agencies to 
index and make their adjudicatory decisions available to the 
public. 

The Freedom of Information Act (FOIA) imposes numerous 
affirmative disclosure obligations on agencies. Under 5 U.S.C. 
552(a)(2), each agency, in accordance with published rules, is 
required to make final adjudicatory decisions and orders i 
available for public inspection and copying unless the materials 
are promptly published and copies are offered for sale. In 
addition, each agency shall maintain and make available for 
public inspection and copying current indexes that provide 
identifying information for the public as to any matter issued, 
adopted, or promulgated. FOIA further mandates that each 
agency shall promptly publish, quarterly or more frequently, 
and distribute copies of each index unless it determines, by 
order published in the Federal Register, that such publication is 
unnecessary and Impracticable. 

Many agencies do, in fact, index and publish or otherwise 
make available to the public their adjudicatory decisions, as 
required under FOIA (e.g., the National Labor Relations Board, 
the Merit Systems Protection Board, the Interstate Commerce 
Commission, the Securities and Exchange Commission). This 
recommendation, then, is addressed to those agencies which 
either entirely fail to Index, publish or make their decisions 
available to the public or fail to do so adequately, whether or not 
they use adjudicatory precedent to pronounce and develop 
agency policy. 

Debate has surrounded consideration of an appropriate test 
for determining which types of adjudicatory decisions are 



^This subsection also covers agency statements of policy and 
interpretations, as well as administrative staff manuals and Instructions to staff 
that affect a member of the public. The Conference has already recommended 
that agency policies that affect the public should be articulated and made 
known to the public to the greatest extent feasible, ACUS Recommendation 71-3 
"Articulation of Agency Policies." See also ACUS Recommendation 70-3, "SEC 
No-Actlon Letters Under Section 4 of the Securities Act of 1933." 



48 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

included In this affirmative disclosure obligation. The Attorney 
General initially expressed the opinion that FOIA requires that 
agencies index only those decisions cited by an agency or relied 
upon as precedent. This limitation, in the view of the Attorney 
General, was derived from both the enforcement provision in the 
statute, which precludes the agency from giving precedential 
effect to matters not indexed, and the legislative history of the 
statute, which indicates that the disclosure provision was 
intended to make available documents having precedential 
significance. The Attorney General also was influenced by the 
impracticality of indexing all agency decisions. 

Application of the affirmative disclosure requirements, 
beyond simply precedential decisions, however, offers several 
advantages. First, if agencies index all significant decisions, 
and not just those decisions deemed to be precedential, agencies 
would be less inclined to be restrictive or one-sided in the 
selection of cases to be accorded precedential effect. Second, 
private parties affected by agency action would be in a better 
position to learn of and influence agency policy. Third, a 
broader application of affirmative disclosure requirements would 
implement the underlying aim of the FOIA indexing 
requirements which is to afford citizens the essential 
information needed to deal effectively and knowledgeably with 
federal agencies and to guard against the development of secret 
law. Lastly, a current index of final decisions may assist 
agencies in developing standards and policies with respect to 
general issues and recurring questions. 

The few cases dealing with the FOIA affirmative disclosure 
obligations have generally read the precedential test broadly. 
They require disclosure not only of decisions that an agency 
considers to be binding but also all decisions that an agency 
retains for general reference and research. The recommended 
approach to the indexing and public availability of final 
decisions focuses less on the binding nature of the precedent 
and more on the value that decisions can have to inform and 
assist the public. 



RECOMMENDATION 89 8. INDEXING OF FOIA & PUBLISHING REQUIREMENTS 49 

RECOMMENDATION 

1 . Indexing of Agency Decisions 

Agencies that do not already do so should compile a 
subject-matter index, of their adjudicatory decisions so as to 
afford citizens inibrmation useful in dealing with the 
agencies and to assist the development of agency standards 
and policies on general issues and recurring questions. 2 

In meeting FOIA indexing requirements, agencies should 
ensure that a subject-matter index is made of their 
decisions and that the index includes all significant 
decisions, whether or not the decisions aie designated as 
precedential. 

2. Level and Scope of Decisions Indexed 

The index should cover the adjudicatory decisions of the 
agency's highest level tribunal. The agency should also 
consider whether to index significant lower level decisions 
that have become final. The adjudicatory decisions 
intended to be covered by this recommendation are those 
made Avith an accompanying written opinion or rationale in 
contested cases after an opporiunity for a hearing at some 
stage of the proceeding. 

3. Index Contents 

Agency Indexes should be designed for effective and 
efficient use. These Indexes should contain sufficient 
information on each indexed decision to identify the major 
issues decided and the location of the case file. Agencies 
should adopt one of the following practices in indexing their 
adjudicatoiy decisions: 

A, Universal Index. Index all final decisions; or 



2ln programs where the agency has established a policy that none of Its 
decisions have precedential effect, the Conference urges that the agency re- 
examine the feasibility of creating a system that accords certcdn decisions 
precedential value to provide guidance about the factors that Influence their 
decisions and to ensure better development of agency policy and standards. See 
ACUS Recommendation 87-7, "A New Role for the Social Security Appeals 
Council," 1 C.F.R 305.87-7. See aLso ACUS Recommendation 71-5, "Procedures 
of the Immigration and Naturalization Service In Respect to Change-of-Status 
Applications." 



50 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

B. Selective Index. Where the volume of decisions 
makes a universal index Impracticable or uninformative, 
selectively index final decisions omitting those decisions 
that are repetitive. The selective index should include all 
significant decisions. Decisions may be significant because 
they are deemed by the agency to be precedential or 
otherwise establish a principle to govern recurring cases 
with similar facts, develop agency policy and exceptions to 
the policy in areas where the law is unsettled, deal with 
important emerging trends, or provide examples of the 
appropriate resolution of major t5rpes of cases not otherwise 
Indexed. 

4. Public Notice of the Index 

Agency indexes should be fully disclosed and readily 
available. Appropriate notice of the existence of 
unpublished decisions should also be given in both the 
agency's FOIA regulations and the procedural or 
substantive regulations governing the specific program. 

5. Computer Technology 

Agencies should explore the use of computer technology 
in order to promote accessibility and reduce costs of 
indexing. 



Recommendation 89-9 

Processing and Review of Visa Denials 



United States consulates around the world complete the 
processing of some nine million applications for immigrant and 
nonimmigrant visas each year. Approximately ninety percent 
are granted: ten percent are denied. Under current practice, the 
only review of a consular official's denial of a visa may be by a 
more senior officer in the consulate, or, on points of law, by the 
Visa Office in the State Department. The Immigration and 
Nationality Act has been read to preclude administrative review, 
and the courts, with a few exceptions, have declined to review 
visa denials. 

Immigrant visas are available to persons with close family 
relationships to U.S. citizens and residents, or with particular 
abilities or skills that are needed but not otherwise available in 
the United States. Nonimmigrant visas are available in a long 
list of classes, ranging from tourists to students to certain types 
of business personnel to diplomats. 

Whatever the visa category or class, there clearly are 
important interests at stake. These interests are not just those 
of the applicants themselves, but also of citizens and residents 
of the United States who are sponsoring the applicant or have 
some other interest in the applicant's presence in the United 
States. These interests warrant a close look at whether initial 
decisions in this important program of mass adjudication 
should be more fuUy reviewable than at present. 

Federal law and State Department regulations give consular 
officers substantial discretion in adjudicating visa applications. 
For example, consular officials exercise absolute discretion in 
determining whether an applicant may be represented by an 
attorney or other qualified representative at the visa application 
interview. Furthermore, although current Department 

regulations, at 22 C.F.R. 41.121(c), require that a denial of a 
visa application be reviewed by a more senior officer, the high 
volume of applications at some posts has resulted in only a 
random sample of denials being reviewed. Review by a senior 
official may also be a problem in single-officer posts. 

Consular posts send a few hundred cases a year presenting 
significant legal issues to the Visa Office of the State 



52 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Department for an advisory opinion that is binding only with 
respect to legal issues. The applicant typically has no notice of 
this proceeding. Such review affects the results in only a small 
number of cases, since most visa denials are based on a factual 
determination. 

Current law has been read by some to limit both 
administrative and Judicial review. Section 104(a) of the INA, 8 
U.S.C. 1104(a), excepts the Secretary of State from the 
administration or enforcement of "those powers, duties and 
functions conferred upon the consular officers relating to the 
granting or refusal of visas." This language has been considered 
by some to preclude the establishment of a more formal review 
mechanism within the State Department. Further, courts have 
generaUy limited the extent of available judicial review. 

The Conference believes that It is important that there be at 
least some level of review of consulgir discretion to deny or grant 
visas. The availability of such review would not only encourage 
consistency and care in the initial adiudicatlon, but would serve 
Interests of fairness and legitimacy. On the other hand, a review 
scheme In this area can be crafted in a fashion that keeps 
procedure to a minimum, takes account of the extremely high 
volume of visa applications, and avoids over-judicialization of 
the process. 

The Recommendation reflects a two-pronged approach to 
administrative review of visa denials, aimed both at improving 
the review at the consular level and at considering the creation 
of a level of centralized administrative review. The suggestions 
directed at the consular offices are intended to encourage quick, 
consistently applied, and cost-effective review that would resolve 
many of the issues on which review might be requested. The 
Recommendation also asks the State Department to study the 
issues, and develop and submit to Congress a proposed process 
for administrative review of consular actions. Tlie Conference 
recognizes that there are currently competing priorities for 
resources that might be required by implementation of this 
Recommendation, but believes that these proposals should be 
implemented as quickly as is feasible under the circumstances. 



RECOMMENDATION 



1. The State Department should adopt a regulation ensuring 
that applicants may be accompanied by an attorney or other 



RECOMMENDATION 89-9, VISA DENIALS PROCESSING 53 

authorized representative during the course of the visa 
application interview process. To the extent practicable, the 
State Department should take steps to reply promptly to 
communications from applicants or authorized representatives 
and to ensure that facilities are available to enable applicants to 
meet with their representatives during the application interview 
process. 

2. The State Department should require consular officers to 
provide brief but explicit written statements of the factual and 
legal bases and reasons for denying a visa application, except 
where reasons of national security or potential adverse effects 
on foreign policy dictate otherwise. 

3. The State Department should modify its regulations to allow 
Visa Office advisory opinions to be made available to applicants 
and their authorized representatives except where national 
security or potential adverse effects on foreign policy dictate 
otherwise. 



4. The State Department should either comply with its 
regulation found at 22 C.F.R 41.121(c) requiring review 
within a consulate of each denial of a visa application, or 
examine alternative systems to review visa denials at consular 
posts. In such a study, the State Department should keep in 
mind the goal of ensuring consistency in visa adjudications and 
consider possible alternatives to address exigencies created by 
busy consular posts, for example, by reviewing random samples 
of visa denials, or selecting for review certain types of denials. 

5. The State Department should, after appropriate study, 
develop and submit for Congressional review a proposed process 
for administrative review of consular visa actions, i 



1 If it Is thought that current law precludes such a State Department study, 
Congress should authorize the State Department to undertake the study. 



Recommendation 89-10 

Improved Use of Medical Personnel in Social 
Security Disability Determinations 



The Social Security Administration annually processes more 
than 1.5 million requests for Disability Insurance Benefits and 
Supplemental Security Income requiring a determination 
whether the claimant is disabled. The Administrative 
Conference has addressed various aspects of the Social Security 
Administration's administrative procedures in earlier 
recommendations. 1 This recommendation focuses more 
specifically on the appropriate use of medical personnel in 
making disability determinations. 

The Social Security Administration (SSA) uses medical 
personnel currently in two ways. First, initial and 
reconsideration determinations are made for SSA by federally 
funded state agencies that use teams composed of one lay 
disability examiner and one medical doctor or psychologist. 2 
Second, medical sources are used to provide evidence of 
disability in individual cases and to explain or elaborate upon 
medical evidence obtained from other sources. Medical sources 
provide evidence relating to individual claims to state agencies 
at the initial decision and reconsideration levels, to 
administrative law judges at the hearing level, and to the 
Appeals Council. Requests can be made to the claimant's 
treating physician or to an independent physician who is asked 
to examine the claimant and report on his or her findings. 
Doctors are asked by some administrative law Judges to explain 
or elaborate upon existing medical evidence; other 
administrative law judges and most state agency personnel do 
not use independent m.edical doctors for these purposes. 
Medical personnel are Involved in the disability determination 



^See Recommendations 78-2 (ALJ hearing stage), 87-6 (state level 
determinations), 87-7 (Appeals Council). 

2For cases Involving mental impairments, Social Security regulations provide 
that either psychologists or psychiatrists may assist In determining disability. 
Accordingly, references to the terms "medical sources," "physicians," and 
"doctors" in these recommendations are intended to Include psychologists used 
in those cases. 



RECOMMENDATION 89-10, MEDICAL DECISIONMAKING BY SSA 55 

process for other federal disability programs as well. Although 
the extent to which they are used varies from program to 
program, programs typically concentrate the use of medical 
personnel at the initial decision stage, as does the Social 
Security Administration. 3 

There is no doubt that medical personnel can offer valuable 
assistance in making disability determinations called for by the 
Social Security Act. Notwithstanding the mixed medical and 
legal content of the Social Security Act's disability standards, 
most disability determinations require the resolution of medical 
issues in one form or another. At the same time, it must be 
recognized that doctors cannot simply apply their general 
medical expertise to the work of determining disability under a 
complex and multi-faceted statutory disability standard. 
Doctors are accustomed to evaluating a person's limitations in 
the context of treatment: they are oriented professionally to 
identify the cause of and resolve limitations, rather than to 
identify limitations and then measure them against stated 
requirements for receipt of benefits. These recommendations 
are intended to help reconcile the needs of the Social Security 
Administration disability determination process for medical 
expertise and the ability of the medical profession to meet those 
needs. 

Medical personnel perform three main functions In current 
practice. First, they assist in developing the medical records on 
which disability decisions are based. Second, they provide 
medical evidence for the record, including medical findings and 
opinions relating to an individual claimant's impairments and 
explanations of other medical evidence already in the record. 
Third, they participate in making disability decisions at the 
initial and reconsideration levels based on the record. 

Each of these functions suggests models for using medical 
decisionmakers in Social Security disability determinations. 
The first model would increase the responsibility of medical 
personnel for compiling all relevant medical evidence. Medical 
personnel would concentrate on evaluating the adequacy of the 
record and following up with requests for clarification and 
additional information from treating and consulting medical 
sources. Medical personnel would also be given specific 
responsibility for assuring that all medical evidence in the 



^While the Conference has examined the other federal disability programs 
and believes that these recommendations hold valuable lessons for the agencies 
administering those programs, these recommendations are addressed solely to 
the Social Security Administration. 



56 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



record is clear and understandable to both medical and non- 
medical decisionmakers. The second model would improve the 
use of doctors as sources for supplying medical data and 
opinions on which disability decisions can be based. This model 
also supports the use of medical personnel to evaluate and 
resolve certain specified medical issues relevant to a claim if, in 
a particular case, there are medical issues that can be identified 
as appropriate for separate decision. The third model would 
make more effective use of medical personnel in a 
decisionmaking role. This model would concentrate medical 
resources at the initial decision level, where a doctor would 
share the responsibility for decisionmaking with a non-medical 
disability examiner. The doctor member of the team would be 
given special responsibility for certain tasks, and would 
undertake a full and independent review of the entire record in 
each case. The expectation is that through open exchange of 
information between the two decisionmakers and a reasonable 
allocation of responsibility based on each member's expertise, 
most disability determinations will be made by consensus. If 
conflicts arise on medical issues, separate medical personnel 
would be given the authority to resolve those conflicts. 

The following recommendations would implement the 
important provisions of each of these models. Implementing 
these recommendations would require greater expenditures for 
medical personnel and related support at the state agencies. 
However, additional costs should be offset by savings resulting 
from elimination of the reconsideration level and reduced 
numbers of administrative and federal court appeals. 



RECOMMENDATION 



A. Improvements at the Initial Decision Level 

The Social Security Administration (SSA) should enhance 
the decisionmaking role of medical personnel at the initial 
decision level. This can be accomplished by improving 
upon the current practice of using two-member teams- 
consisting of a medical member who is a licensed physician 
or psychologist and a non-medical member who is a 
disability examiner- -to determine disability, as follows: 

1 . Responsibility for developing medical evidence. SSA 
should ensure that the medical member of the team is given 



RECOMMENDATION 89-10, MEDICAL DECISIONMAKING BY SSA 57 

primary responsibility for developing the medical evidence* 
in the record. 

(a) Staff and resources should be allocated so as to 
assure that a complete record of all evidence relevant to 
a disability claim is obtained before an initial decision 
is made on the claim. 

(b) Specially trained support staff, including nurses 
and non-medical personnel, should be made available 
to assist the medical member in developing the medical 
evidence. 

(c) The medical member should, whenever possible, 
be assigned direct responsibility for evaluating the 
adequacy of reports from physicians and for following 
up with requests for clarification or additional 
Information from these sources. 

2. Identifying and deciding discrete medical issues. 
SSA should develop a list of discrete Issues raised by the 
applicable disability standards that may arise in Individual 
claims and that are appropriate for decision by medical 
staff. The medical member of the team assigned to a claim 
should be made responsible for identifying any such 
discrete issues raised in the claim, developing aU evidence 
relevant to the issue, and reaching a decision on that issue. 

3. Resolving medical conflicts. SSA should ensure that 
medical personnel are used to resolve any conflicts on 
medical Issues that arise In the course of team evaluations 
of disability at the initial decision level. 

(a) Senior medical staff should be given the 
authority to review claims where the team members are 
unable to agree and to recommend further action, 
including the development of additional medical 
evidence, to resolve the conflict. 

(b) ff the conflict persists, the state agency's 
medical personnel should assume primary 
responsibility for evaluating the record with respect to 
the medical issues and for making a determination 
based on that record. 



^"Medical evidence" Includes (1) medical findings and opinions relating to an 
Individual claimant's impairments, (2) other evidence, Including subjective 
symptoms, that is relevant to determining the existence or severity of the 
claimant's condition, and (3) explanations of other medical evidence already In 
the record. The recommendations' focus on development of medical evidence is 
not intended to minimize the importance of the development of other evidence, 
including vocational evidence. 



58 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

(c) As part of this process. Independent medical 
experts, or panels of experts, should be identified and 
retained for use as examining and non-examining 
consultants, as appropriate. 

4. Notice of deficiencies tn medical evidence. SSA 
should require that claimants be informed specifically of 
any deficiencies In the medical evidence that could lead to 
an adverse determination before the initial decision is 
made. 

(a) This notice should be prepared by the medical 
member of the team, should clearly explain any 
deficiency in the medical evidence, and should 
encourage the claimant to provide additional 
information and explanation, as needed. This notice 
should also state that the agency will assist claimants 
in obtaining this information when they are unable to 
do so on their own due to financial or other constraints. 

(b) As part of this process, either the claimant or 
the medical member should have the authority to 
initiate a face-to-face Interview. 

5. E^isuring quality of evidence. SSA should take steps 
to improve the quality of evidence provided by medical 
sources for disability adjudications. 

(a) Guidelines should be established that identify 
priorities for the use of treating physicians, examining 
physicians and non-examining physicians, including 
specialists, for these purposes. 

(b) Selection and evaluation of physicians asked to 
provide medical information should be performed by 
medical personnel independent from the agency staff 
responsible for making disability decisions and should 
be supported by a system for quality control covering 
both the selection of physicians and the reports 
submitted. 

(c) Physicians asked to provide medical information 
should be adequately compensated and should be 
provided with instructions as to applicable agency 
standards. 

(d) Medical personnel should be able, when 
appropriate, to consult with specialists before ordering 
examinations or tests. 

(e) All contacts with medical sources relating to the 
determination of disability for a particular claim should 
be documented routinely in writing and included in the 
record. SSA should ensure that claimants are provided 



I 



RECOMMENDATION 89-10, MEDICAL DECISIONMAKING BY SSA 59 

a copy of any reports prior to issuance of the decision 
and accorded an opportunity to object and rebut 
appropriately. 

6. Training and supervision of medical personnel SSA 
should ensure that all medical personnel are trained fully 
on legal and program issues and work under the 
supervision of the chief medical officer in the state agency. 
SSA should also ensure that medical staff act in accordance 
with the rules established by the Social Security Act and 
relevant federal court decisions, including the requirement 
to obtain and give appropriate weight to the opinions of 
claimants' treating physicians, in performing the functions 
described in paragraphs 2, 3(b), and 5(a). 

B. Reconsideration 

7. Elimination of Reconsideration. SSA should seek to 
concentrate the efforts of the disability determination team 
on a single initial decision process, as outlined in these 
recommendations. Together with implementation of these 
recommendations, the separate reconsideration stage 
should be eliminated. 

C. Appeal Level 

8. ALJ use of medical experts. SSA should encourage 
Its administrative law judges to call on an Independent 
medical expert in appropriate cases to assess the need for 
any additional medical evidence and to explain or clarify 
medical evidence in the record. 5 SSA should make clear by 
regulation that a medical expert's evidence can be 
presented orally or in writing. The regulations should also 
provide that claimants are notified of the inclusion of an 
expert's report in the record and should assure that 
claimants' rights to object to the inclusion of the report, 
submit rebuttal evidence, and cross-examine the expert are 
not abridged. The regulations should also provide that all 
information and opinions provided by medical experts must 
be included in the record. 



5SSA should also ensure that Its ALJs receive appropriate training on 
medical Issues relevant to their decisional responslbllles. 



statement 14 

Mass Decisionmaking Programs: The Alien 
Legalization Experience. 



The Alien Legalization Program, authorized under the 
Immigration Reform and Control Act of 1986, created a program 
to allow certain aliens present illegally in the United States to 
convert their status to that of a legal resident. The program is 
administered by the Immigration and Naturalization Service and 
has a short and defined lifetime, which is heading toward 
completion. The legalization program contains two separate 
pieces: the "section 245A" program, for aliens who have been in 
the country since January 1, 1982; and the "special agricultural 
worker" program, for alien farmworkers who worked in specified 
agricultural employment for at least 90 days during a specified 
period. The application period for the "section 245A" program 
ran from May 5, 1987 through May 4, 1988. The application for 
the SAW program ran from June 1, 1987 through November 30, 
1988. Although the INS has acted on most of the cases, some 
are likely to remain pending for months. Moreover, as described 
below, the second phase of the process is still ongoing. 

The Administrative Conference has studied the INS's 
processing of alien legalization applications from the 
standpoints of what improvements can be made by the INS in 
the remaining phases of the legalization program itself, and 
what lessons can be applied in future mass decisionmaking 
programs by the INS or by other agencies. This Statement does 
not address the merits of litigation over the regulatory ground 
rules of the program, but only the procedures for handling the 
applications themselves. 



STATEMENT 14. ALIEN LEGALIZATION 61 

DESCRIPTION OF THE PROCESS 



The Allen Legalization Program has been administered by the 
INS using a framework of local Legalization Offices (LOs) (107 of 
them across the country) and four Regional Processing Facilities 
(RPFs) to process the more than three million applications for 
legalization that were received. Applications are filed with the 
LOs, where Interviews are conducted and recommendations for 
action are made. The files are then sent to a central processing 
center in London, Kentucky, following which they are forwarded 
to one of the four RPFs across the country. The RPFs make the 
determinations on the applications, based on the fUe material. 
Appeal of a denial of legalization status is available to the 
Legalization Appeals Unit (LAU) in Washington, D.C., and is 
based on the "administrative record." 8 U.S.C. 1255a(ll(3)(B). 

The legalization process has two phases. The first phase is to 
determine whether an applicant qualifies for legalization. The 
second phase, which applies only to qualified section 245A 
applicants, involves a determination whether they qualify for 
permanent resident alien status. (In the agricultural program, 
permanent resident alien status is automatic.) The initial 
application period for both programs is now closed, and the INS 
has processed a large number, although not all, of those 
applications. 

The "remote decisionmaking" system employed in the Alien 
Legalization Program Involves decisionmaking at the four RPFs 
rather than at local levels. This system has provided more 
consistency in decisions than other, more decentralized systems 
within the INS. Having the determination made by a person 
removed from the individual who actually interviewed the 
applicant reduced the potential subjectivity of decisionmakers. 
Because interviewers at the local levels knew that their files 
would be examined in virtually every case, the quality of work in 
the files has been relatively high. There are also suggestions 
that the system of regional facilities promoted cost-efficiency. 

The RPF remote decisionmaking system, however, has not 
been without problems. The elimination of direct contact 
between the decisionmaker and the applicant, which helps 
eliminate bias or prejudice, also eliminates the opportunity for 
first-hand credibility determinations by the ultimate 
decisionmaker. The RPFs also have been very isolated from the 



62 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

public, making it difficult for applicants or their representatives 
to acquire information about the status of cases, among other 
things. The RPFs have had only limited access to legal advice 
from INS attorneys, requiring them in at least one facility to 
seek advice from the LAU, which is the appellate unit that 
reviews appeals of their decisions. There have been some 
difficulties relating to the Interaction between the LOs and the 
RPFs, and relating to the provision of adequate notice to 
applicants at different stages of the program. There also has 
been reported an unexpectedly high incidence of fraudulent 
applications, particularly in the agricultural worker program, 
which the RPFs were not equipped to handle. 

Overall, however, the system for deciding legalization 
applications appears to be working fairly well. The results 
among the regions have been quite consistent so far. The INS 
has been able to process large numbers of the applications, 
particularly in the first phase of the section 245A program. 
Moreover, the INS has recognized many of the problems, and 
taken steps to address them. 

From the INS experience, it is possible to derive some 
lessons, not only for other INS programs in the future, but for 
other agencies that may consider using a system of remote 
decisionmaking for a large volume of cases. There are also a 
number of lessons that can be drawn about the Alien 
Legalization Program in particular. While that program is 
nearing completion, to the extent that these comments refer to 
elements of the program that are ongoing, the Conference 
encourages the INS to implement the suggestions to the extent 
possible and to the extent that it has not already done so. 



CONCLUSIONS 



A The Conference encourages agencies, including the INS, to 
consider using remote decisionmaking where there is a large 
volume of cases to be decided on the basis of objectively 
verifiable information within a written file, particularly where 
bias, prejudice or other subjectivity may be a significant 
problem. This system appears to promote consistency among 
decisionmakers, perhaps because of the more centralized nature 
of the process, and the fact that it is easier for a small group of 
managers to confer on a regular basis about the decisionmaking 



STATEMENT 14, ALIEN LEGALIZATION 63 

process. Agencies should, in Implementing such a system, 
consider the following suggestions: 

1. In a remote decisionmaking system in which a file 
prepared on a local level will be the basis for 
decisionmaking, there must be clear guidelines as to what 
are the necessary contents of the file, and the use of 
standardized forms and checklists is encouraged. 

2. Where the local office is making a preliminary 
recommendation to the remote decisionmaking center, the 
local office should be instructed to provide an explanation 
for its recommended decision that is sufficient to provide 
the remote center with the maximum benefit of the local 
office's information and interaction with the applicant or 
other interested person whose case is being determined 
(hereinafter "applicant"). Appropriate forms or formats (e.g., 
computer entry) should be designed to ensure that the 
necessary information is transmitted in a uniform manner 
to assist in review and retrieval. 

3. It is important that adequate information be 
available to the applicants and their representatives 
concerning their cases. To facilitate this, case tracking 
sjTstems capable of responding to inquiries should be 
developed. Applicants should be clearly Informed of the 
process relating to their cases, and be given adequate notice 
of each step that requires or provides an opportunity for 
action or participation on their part. 

4. To avoid the appearance or actuality of conflict, it is 
Important that the remote decisionmaking centers have 
adequate access to legal advice relating to the merits of 
matters before them from agency legal staff other than from 
the appellate unit, if any, that reviews appeals of their 
decisions. 

5. Ensuring input from local personnel on credibility 
issues is of pzirticular importance. It may be useful to 
consider ways of videotaping or otherwise recording 
interviews where the applicant's credibility is at issue. 

6. Where helpful, the decisionmakers at remote 
decisionmaking centers should be provided the opportunity 
to work for a short period of time at a local office, giving 
them first-hand experience in interviewing applicants, in 
order to provide them a better sense of the implications of 
the information they receive. 

7. The remote decisionmaking centers' managers 
should consult with each other regularly on substantive 



64 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

and procedural matters, in order to ensure that their 
actions are consistent. 

8. It is important that the system be able to provide 
new or amended instructions to the local offices quickly, so 
that consistency can be maintained and the benefits of 
experience transmitted. 

B. With respect to the Alien Legalization Program specifically, 
the Conference encourages the INS to implement or continue to 
implement the following suggestions to the extent that they refer 
to elements of the program that are ongoing: 

1. Action should be taken to ensure that information 
concerning individual applications for legalization is readily 
available to the applicant or his or her representative. INS 
should: 

a) Establish a method of tracking cases that would 
enable applicants or their representatives to obtain 
information expeditiously concerning the status of their 
cases; 

b) Ensure that applicants are clearly informed 
concerning all of the steps required to complete the 
legalization process, both initially and as actions are 
taken on their applications, particularly where further 
action on their part may be called for; 

c) Recognize the need to extend completion times 
where applicants are unable to fulfill educational 
requirements due to a shortage of educational facilities 
or programs; 

d) Supplement existing mail communication with 
applicants, including through the use of broadcast 
media. Particular attention should be paid to the 
mobility of many applicants, as well as to the fact that 
applicants have not always understood the multiple- 
step nature of the process, and thus have not realized 
that INS may be seeking to communicate with them. 

2. INS should ensure that RPFs have sufficient access 
to legal advice from INS attorneys; however, the RPFs 
should not seek legal advice from the Legalization Appeals 
Unit (LAU), the office that reviews appeals of RPF decisions. 

3. INS should consider making the following 
modifications in the way applications are processed by the 
Legalization Offices (LOs) and RPFs: 

a) Including in the recommendation forms used by 
the LOs an additional option, "recommended scrutiny". 



STATEMENT 14, ALIEN LEGALIZATION 65 

in order to assure that the RPF reviews the application, 
without having to recommend denial for lack of another 
option. 

b) Arranging for the following In cases where fraud 
is suspected: 

i) reinterviewing applicants whose cases 
have not been decided: in the case of SAW 
applicants, such interviews should be done by 
specially-trained decisionmakers with knowledge 
of agriculture. 

ii) using video cameras to tape remaining 
interviews, in order to have a record on which 
credibility can be better assessed. 

iii) assigning remaining cases at the RPFs to 
specially trained teams of decisionmakers: in the 
case of SAW applicants, such decisionmakers 
should be trained to review agricultural cases 
and the types of fraud that may appear in such 
cases. 

c) Considering whether LO personnel should have 
more direct impact on legalization decisions, since they 
are the ones who actually have the opportunity to 
assess the credibility of applicants. 



statement 15 

Procedures for Resolving Federal Personnel 
Disputes 



In 1978, Congress enacted the most comprehensive revision 
of the federal civil service laws since the Pendleton Act of 1883. 
The Civil Service Reform Act (CSRA) of 1978 created new 
institutions and processes for personnel management, including 
a substantially restructured system for considering complaints, 
grievances and appeals filed by federal employees in response to 
personnel actions, i Personnel disputes typically involve such 
matters as removals from the service, reductions in pay or 
grade, suspensions, furloughs, promotion and award decisions, 
and reductions-in-force. 

Congress' 1978 revision of the framework for considering 
employee complaints, grievances and appeals was in part 
prompted by the confusing, complex and time-consuming 
nature of the existing procedures. Moreover, the procedures 
were seen as intimidating to both managers and employees and. 



Ipive Institutions share in the administration of the CSRA: 

1. Office of Personnel Management (OPM), which is vested with the 
general authority to execute, administer and enforce the laws and 
rules governing the civil service except with respect to those functions 
vested in other agencies; 

2.Merit Systems Protection Board (MSPB). which is an 
independent adjudicatory body charged with hearing and deciding 
employee apjjeals from agency personnel actions of various kinds 
Including major performance and disciplinary actions; 

S.Offlce of the Special Counsel (OSC), which is charged with the 
investigation and prosecution before the MSPB of requests for 
corrective and disciplinary action when prohibited personnel 
practices have allegedly occurred; 

4. Equal Employment Opportunity Commission (EEOC), which 
enforces various anti-discrimination statutes as they apply to both 
public and private employment, Including Title VII of the Civil Rights 
Act of 1964, the Age Discrimination in Employment Act, and the 
Rehabilitation Act; and 

5. Federal Labor Relations Authority (FLRA), which is an 
independent agency charged with hearing and deciding complaints of 
unfair labor practices and reviewing arbitration awards on matters 
falling within the scope of negotiated grievance procedures 
established by collective bargaining agreements. 



I 



STATEMENT 15, FEDERAL PERSONNEL DISPUTES 67 

therefore, likely to deter managers from taking appropriate 
personnel action and to discourage employees from pursuing 
available avenues to vindicate their rights. There was also a 
perception that the Civil Service Commission was 
unsympathetic to employee discrimination cases and to unions. 

The statutory framework created in 1978, with its new 
forums for adjudicating employee claims, was no less complex 
and has continued to produce costs, delays, and confusion, 
accompanied by substantial resource commitments. Close 
scrutiny discloses that this complexity was in large part created 
by the tension Inherent in the simultaneous pursuit of three 
important legislative goals: adequate protection of employees 
from discrimination; 2 uniformity in federal personnel 
management; and solidification and expansion of the place of 
arbitration as a mechanism to resolve federal employee 
grievances. 

As a general principle, procedures for handling federal 
employee disputes should be as simple and fair as feasible. 
Fairness must include a concern for the timely, final resolution 
of claims. At the same time, however desirable procedural 
simplification may be, federal employees and managers alike 
should have confidence in the processes for adjudicating 
personnel disputes. This is especially true when a federal 
employee contests a personnel action on the grounds of 
discrimination, since nothing is more basic than the right to be 
free of invidious discrimination. 

To address these matters, the Administrative Conference 
commissioned a consultant's report. 3 The report describes the 
historical background to the 1978 reforms, along with the 
institutional and procedural framework created by the CSRA. It 
examines the operation of the system and makes specific 
recommendations for legislative and administrative changes. 
The Conference's Committee on Governmental Processes has 
held numerous meetings to consider the report, has provided 
opportunity for public participation in those meetings, and has 
invited public comment on various procedural changes that 
might be recommended. 

This process has demonstrated the continuing importance of 
employee and management concerns in addressing the 



^'Discrimination" refers to discrimination on the basis of race, color, 
religion, sex, national origin, age, and handicapping condition. 

^William V. Luneburg, The Federal Personnel Complaint. Appeal, and 
Grievance Systems: A Structural Overview and Proposed Revisions, 78 KY. L. J. 1 
(1989). 



68 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

procedural questions that are the subject of the study. The 
intricate web of processes, consciously designed by Congress in 
its enactment of the CSRA. resulted in part from substantive 
policy and value judgments that transcend the domain of 
administrative procedure. The complex framework has been 
perceived by employees and their representatives as providing 
necessary protection against possible arbitrary employer action, 
particularly in cases where discrimination is alleged. As a 
result, employees may perceive proposed procedural 
simplification as a potential threat to substantive rights. 

For this reason, whether to retain or modily the present 
procedures must be resolved largely through the legislative 
process. The experience of the past decade, as described in the 
consultant's report, has persuaded the Administrative 
Conference that the time is ripe for fresh consideration by 
Congress of the problems that gave rise to the CSRA and to the 
issues that have developed subsequently. In fact. Congress has 
started to re-examine selected aspects of the system, including 
equal employment opportunity. In addition, the EEOC has 
issued a notice of proposed rulemaking to change substantially 
the federal sector discrimination complaint process."* 

This Statement describes certain important issues that we 
believe should be addressed as part of a comprehensive 
congressional review of the CSRA processes. These include: 

(i) whether there should be one rather than the present 

two opportunities for a full scale trial-type proceeding in 

cases raising claims of discrimination: 

(U) whether to retain the existing complex structure for 

administrative adjudication of "mixed" cases, i.e., those in 

which the employee alleges that unlawful discrimination 

was the basis for an adverse personnel action; 

(tii) whether the government should have the right to 

appeal to the courts in cases involving claims of 

discrimination; 

(iv) whether employees or their unions should be 

permitted to seek judicial enforcement of arbitral awards; 

and 

(v) whether the functioning of existing adjudicatory 

frameworks— grievance systems of individual agencies, 

agency complaint processes for discrimination cases, the 

MSPB process and the various negotiated grievance 

processes— could and should be improved. 



4See "Federal Sector Equal Employment Opportunity" (29 CFR Part 1614), 
54 Fed. Reg. 45747 (October 31. 1989). 



STATEMENT 15, FEDERAL PERSONNEL DISPUTES 69 



These questions cannot and should not be examined in 
isolation, as any review of the procedures of the CSRA should 
recognize the integrated nature of the complex processes of the 
Act. Whether or not congressional re-examination takes place, 
this Statement also suggests that the Office of Personnel 
Management and agencies scrutinize and, where appropriate, 
seek ways to improve the patterns of communication within 
agencies and to employees regarding the workings of the 
complaint, grievance and appeal system. The Conference 
encourages the creative use of alternative means of dispute 
resolution in resolving employee claims. 5 

A. Need for Comprehensive Overview. 

In considering the issues posed below. Congress should 
take into account the fact that the intricate federal 
personnel scheme consists of numerous interrelated 
processes and institutions. Accordingly, statutory changes 
should be enacted only after careful consideration of their 
potential direct and indirect consequences throughout the 
system for resolving employee complaints, grievances and 
appeals. 

B. Cases Raising Claims of Discrimination. 

1 . The Number of Hearings in Mixed Cases. "Mixed" 
cases are controversies involving personnel actions 
appealable to the MSPB, where an employee alleges that 
unlawful discrimination was a basis for the agency's action. 
These cases are "mixed" in that they involve issues both of 
civil service law and employment discrimination. Under 
present law federal employees, unlike employees in the 
private sector, on allegations of emplojnnent discrimination 
in such cases, have a right to a full trial-type administrative 
hearing (at the MSPB) and, if unsuccessful, may obtain a 
second de novo adjudicatory hearing in United States 
District Court. The question for congressional 

consideration Is whether an employee who has chosen a 
trial-type administrative hearing in a mixed case, and has 
then been afforded the opportunity for a full evidentiary 
hearing on the record on all relevant matters, should be 
entitled to a judicial trial de novo or should be limited to 



^See Recommendation 86-3, Agencies' Use of Alternative Means of Dispute 
Resolution. 1 CFR 305.86-3. 



70 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

judicial review in a court of appeals on the administrative 
record under the substantial evidence test— the customary 
venue and scope of review of formal administrative 
adjudications. 

2. The Process for Deciding Mixed Cases. The statutory 
procedure for mixed cases provides the opportunity for 
sequential consideration by the MSPB and the EEOC, with 
the possibility of further proceedings before a Special Panel 
consisting of one member of the MSPB, one member of the 
EEOC, and a chairman appointed by the President for a 
term of six years with the advice and consent of the Senate. 
Through these provisions, the CSRA embodies Congress' 
effort to bring to bear both the MSPB's expertise in general 
issues of personnel law and the EEOC's expertise with 
discrimination issues. This complex procedure is intended 
to achieve a balanced use of both agencies' expertise. 

In most mixed cases presented to the EEOC, the 
Commission has accepted the MSPB decision and, where it 
has not, the MSPB has generally deferred to the 
Commission's disposition of these cases. During the first 
ten years of the CSRA, there have been only three Special 
Panel proceedings, indicating that the two agencies have 
failed to reach an accommodation in only a very limited 
number of cases. E^/^en where a Special Panel was 
unnecessziry, however, sequential consideration by the 
MSPB and the EEOC has required significant time and 
effort and has at times led to confusion on the part of 
agencies and claimants. The question for congressional 
consideration is whether experience confirms the 
desirability of dual agency participation in mixed cases or 
whether that experience suggests that consistently fair and 
more timely results may be achieved through a less complex 
process. 

3. Government's Right to Appeal. Under existing law 
the government may petition for judicial review of certain 
administrative and arbitral decisions in personnel cases 
that present significant issues of civil service law. The 
availability of such review is either uncertain or non- 
existent when those decisions implicate issues of 
discrimination law.6 There is disagreement over the 
desirability of permitting the government to seek a judicial 
resolution of legal issues that have been decided adversely 
to the government at the administrative level. 



1 



^See. e.g., Moore v. Devlne. 780 F.2d 1559 (1 1th Clr. 1986). 



STATEMENT 15, FEDERAL PERSONNEL DISPUTES 71 

The question for congressional consideration is whether 
in the interest of consistency the government should have 
the right to judicial review (or at least the authority to 
petition for it) in cases where the Director of OPM 
determines that an administrative or arbitral decision 
constitutes an incorrect interpretation of civil service or 
discrimination law that may have a substantial impact. The 
primary countervailing considerations are the increased 
potential for further delay in final resolution of such cases 
and whether the executive branch should be authorized to 
seek judicial resolution of disputes between two of its 
agencies, as would be the case if OPM were permitted to 
appeal from adverse determinations of the EEOC. 

C. Enforcement of Arbitration Awards. 

Although federal employee grievance arbitration awards 
are intended to be binding, federal agencies do not always 
comply with them. Under the CSRA, the only mechanism 
available to a federal employee seeking to enforce an 
arbitral award is an unfair labor practice proceeding before 
the FLRA. Such a proceeding requires filing of a complaint 
by the Authority's General Counsel and a hearing before the 
FLRA, followed by judicial review and enforcement of the 
resulting FLRA order. There is substantial dispute over the 
question whether an employing agency may raise any 
defense to an enforcement action other than compliance 
with the arbitral award. 

Questions for congressional consideration include 
whether to permit an employee or union to seek direct 
judicial enforcement of federal sector grievance arbitration 
awards, where such enforcement should take place, and 
what issues should be open for litigation in the chosen 
forum. A central issue is whether permitting such direct 
enforcement actions would increase or decrease the 
likelihood of voluntary agency compliance with arbitration 
awards. 

D. Need to Examine Existing Adjudicatory Systems. 

1. Agency Administrative Grievance Systems. Agency 
administrative grievance systems have taken on increasing 
importance as the only avenue of redress^ available to an 
individual employee in cases not falling within MSPB or 



'7See, e.g., Harrison v. Bowen, 815 F.2d 1505 (D.C. Cir. 1987). 



72 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

EEOC jurisdiction or within the scope of a negotiated 
grievance procedure. The Office of Personnel Management 
and other agencies should ensure that these systems in fact 
provide fair, efficient, and timely resolution of workplace 
disputes, and further studies of the structure and usage of 
these systems may be needed. 8 In reviewing these 
processes, agencies should give careful consideration to 
alternative means of dispute resolution. 

2. Agency Complaint Processes In Discrimination Cases. 
Over the years, criticism has been leveled at the quality and 
timeliness of agency disposition of discrimination 
complaints. The EEOC and individual employing agencies 
should continue to take steps to ensure adequate and 
expeditious investigation and timely, final resolution of 
such claims. Experimentation with, and incorporation of, 
alternative dispute resolution techniques may be useful. 

3. MSPB and Negotiated Grievance Systems. The MSPB 
process and the various negotiated grievance processes 
dispose of large numbers of significant employee claims. 
Concerns have been expressed regarding^ certain aspects of 
their operation, including the familiarity of arbitrators with 
federal personnel law and the need for increasing the 
statutory Independence of the MSPB's administrative 
Judges. These concerns may warrant further study. 

E. Better Communication Within Agencies and to 
Employees. 

Multiple avenues for relief present the danger of parallel, 
duplicative proceedings. Agencies should take steps to 
ensure that there are consistent patterns of communication 
among their personnel, labor relations and equal 
employment opportunity staffs to reduce the likelihood of 
parallel, duplicative proceedings In personnel actions for 
which such duplication is not permitted by law. 

To use the personnel grievance system effectively, 
employees must have clear and reliable Information about 
options available to them and the consequences of choosing 
one route over another. The Office of Personnel 
Management and other agencies should ensure that there 
exist readily available and easily understandable written or 



®OPM has recently reviewed the systems established by agencies and has 
published a study entitled A SURVEY OF AGENCY ADMINISTRATIVE 
GRIEVANCE SYSTEMS (June 1989). 



MEDICARE PEER REVIEW 73 

Other materials that inform federal employees of their 
complaint, grievance and appeal options and the 
consequences of their choice of one complaint, grievance or 
appellate route over another. 



REPORTS 

FOR 
RECOMMENDATIONS 89-1 THROUGH 89-10 

AND 
STATEMENTS 14 AND 15 



Administrative Law Issues Involving the 

Medicare Utilization and Quality Control 

Peer Review Organization (PRO) Program: 

Analysis and Recommendations 

Timothy Stoltzris Jost* 
I. Introduction 

M. The PRO Program 

III. Publication of PRO Program Poucies and Procedures 

IV. Issues Concerning PRO Crtteria, Norms, and Standards 

A. Are PRO Norms and Criteria Sufficiently Definite? 

B . Local or National Norms 

C. Consideration of Social Criteria 

D. Formulation and Publication of Criteria and Standards 
V. The pro Quality Assurance and Sanction Program 

A. Introduction 

B . Criticisms of the PRO Sanction Process 

C. The PRO Quality Assurance and Sanction Process Described 

D. Major Issues in the PRO Sanction Process 

1 . Does the Constitution Require a Pre-exclusion Hearing? 

2. Notice 

3. Confidentiality and Disclosure Issues 

4. Burden of Proof and Weight of Evidence Issues 

5. The "Unwilling or Unable" Requirement 

6. Money Penalties 

7. Bias 

E. Two Alternatives for Improving the PRO Sanction Process 

1 . Transfer Sanction Authority to the Inspector General and Pro- 
vide Pretermination Hearings 

2. Retain Current System, but Streamline and Assure Due Process 
VI. Denials of Payment for Substandard Care 

VII. PRO Investigations of Benerciary Complaints 
VIII. Hospital Notices of Noncoverage 

* Professor of Law and Hospital and Health Services Administration, The Ohio State University. B.A. 1970. 
University of California. Santa Cruz; J.D. 1975. University of Chicago. This Article is denved from a research paper 
submined under a contract with the Administrative Conference of the United Stales. It represents solely the views of the 
author, and its conclusions have not been approved by the Administrative Conference. 

The author expresses his gratitude to the more than 80 experts on the PRO program who were interviewed in 
picparation for this Article. Special thanks are due to representatives from 12 PROs who cooperated and gave in-depth 
interviews. While singling out any of the many individuals who assisted in this study risks slighting others who also 
contributed much, the author must acknowledge the substantial contnburions of Alan Kaplan. Alice Gosfleld. Nancy 
Miller, and Sanford Teplilzky. who offered extensive comments on drafts: Jeffrey Lubbers and Philomena Dane, editors: 
Randal Bannger, icsearch assistant: and Michele Whetzel-Newton. secretary. 



78 TIMOTHY STOLZFUS JOST 

IX. Utilization Review Denials. Reconsiderations, and Appeals 

X. PRO Data CoNnDEmiALmr and Disclosure 

XI. Conclusion 

I. Introduction 
The Medicare Utilization and Quality Control Peer Review Organization (PRO) 
program is an administrative law cunosity. The PRO program is the federal 
government's pnmary tool for assuring that services provided to Medicare benefi- 
ciaries are medically necessary, are of a quality that meets professionally recognized 
standards of health care, and are provided in an appropnate setting.' It both protects 
the health and safety of Medicare beneficiaries and controls the cost of the Medicare 
program. 

The power of PROs over Medicare providers, practitioners, and beneficiaries is 
sweeping. If a PRO determines that medical services do not meet utilization or quality 
standards, it may retrospectively deny Medicare payment for those services.^ A PRO 
may also deny payment prospectively for some prescnbed procedures, effectively 
blocking a beneficiary from receiving those services unless the beneficiary can 
independently afford them.' It may also recommend to the Office of the Inspector 
General (OIG) of the Department of Health and Human Services (HHS) that a 
provider or practitioner* be fined or excluded from receiving payment under the 
Medicare program. As a practical matter, exclusion from Medicare may make it 
impossible for a physician to practice; thus the PRO's power over physicians is nearly 
as great as that of state licensure boards. 

More striking than the scope of the PRO's authority is the fact that in many 
instances PRO decisions are either not reviewable or are reviewable only after they 
have been implemented. A hospital or physician, for example, cannot in most cases 
obtain independent review of the decision of a PRO to deny payment for a claim from 
either an Administrative Law Judge (ALI) or a court — the PRO's decision is final.' 
PRO-initiated sanctions and penalties assessed against providers and practitioners are 
usually not reviewable until months after they have been implemented.^ 

From an administrative law perspective, however, the most striking feature of 
PROs is that, despite their substantial, often unreviewable power, they are private 
entities that provide services for the federal government on a contractual basis. ^ The 

1, Social Sccumy Act. 42 U.S.C. § 1320c-3(aMI) (1982 & Supp. IV 1986) 

2. Id § 1320c-3(a)(2). 

3 Health Care Fin Admin. U.S. Dep r of Heal. h & Human Servs Third Score of Work § X. at 26 ( 1 937^ 
I hereinafter Third Score of Work). The Third Scope o) Work is the request for proposals governing PRO contracts dunng 
the current contract cycle: for further explanation, see mfra text following note 56 

4 UmJer the Medicate law. a provider is an institutional health care entity, such as a hospital or nursing home. 
42 use § I395x(u) A practitioner is a physician or other individual who provides health care. 

5. See infra text accompanying notes 423-27. 

6. 42 use. S l320c-5(bH2)(l982);42C.F.R. §S 1004 lOCKb). 130(a)(3) (1987). A recent draft report of the 
HHS OIG notes that on average it takes 15 months from the date of a PRO recommendation to the OIG to compleuon 
of the appeal. Office of Inspector General. U.S. Dep't of Healhi & Human Servs.. The Utilization and (Juauty 
Control PEtR Review Organization (PRO) Program: Sanction AcTivrriES (draft) 16 (Mar. 1988) [hereinafter OIG 
Sanction Refo«t| There is. however, a special exception for certain raral practitioners, who may receive a pre-exclusion 
AU heanng. 42 U S C. § l320c-5(bH5) (1982). 

7 42 use. §§ I320C-I to -2(b) (1982 & Supp. IV 1986). 



MEDICARE PEER REVIEW 79 



pnvate nature of PROs is intentional and based on the philosophy of medical peer 
review. From the beginning, PROs (like the Professional Standards Review Organi- 
zations before them) were established so that pnvate doctors participating m the 
Medicare program could have their work reviewed by other private doctors, with a 
minimum of interference from the federal government. The quasi-pnvate nature of 
PROs is also reflected in the scope of their activities. TTiough many PROs were 
formed explicitly to perform Medicare review, they are not limited to this function; 
indeed federal -law encourages them also to perform review for private and other 
public entities.* A recent study found that, in fact, 73% of PROs conduct review for 
business entities and 68% for state Medicaid programs.' Several spent less than 50% 
of their time on Medicare-related activities. 

As private entities, PROs are not generally subject to the statutes governing 
federal administrative law. The court in Public Citizen Health Research Group v. 
HHS^^ concluded that Professional Standards Review Organizations (PSROs). the 
predecessors of PROs, were private independent contractors and not federal agencies 
for purposes of the Freedom of Information Act (FOIA). A more recent case held that 
PROs are not federal agencies for purposes of the attorneys fees provisions of the 
Equal Access to Justice Act." Presumably PROs are also not federal agencies for 
purposes of the Administrative Procedure Act (APA).'^ 

This is not to say that PROs are unfettered by administrative law. The federal 
PRO statute, regulations, and Manual specify procedures to be followed by the PROs 
for rulemaking, adjudication, and data disclosure. Indeed, as the importance of the 
PRO program has grown, these requirements have tended to proliferate, providing 
increasingly more protection for those affected by the PROs, but also leaving the law 
affecting PROs in an increasingly confused state. Moreover, in canying out their 
federal review activities, PROs are, as will be discussed below, federal actors subject 
to the due process requirements of the Constitution. '^ There is, therefore, a 
substantial body of statutory, regulatory, and constitutional law that dictates how the 
PROs should conduct themselves in relating to the beneficiaries, providers, and 
practitioners they regulate. 



8. Id. § 1320c-3(aMII) (1982). See Hastings. Legal luues Raised by Pnvate Review Activities of Medical 
PeerRe\iew Organizations. 8 J. Health Pol. Poly & L. 293 (1983) (an excellent anicle considenng the PRO private 
review side, which is not considered in this Anicle). 

9. Office of Inspktoh General. U.S. Dept of Health & Human Sekvs. , The Utilization and Quautt Conthol 
Peer Review Organization (PRO) <^ogram: An Exfijoratiok of Program Effectiveness (draft) l (June 1988) 
(hereinafter GIG Effectiveness JIeport). 

10. 668 F 2d S37 (1981). This holding is leafTiimed by the more recently adopted PRO statute, which explicitly 
sutes thai PROs are not federal agencies under the FOIA. 42 U.S.C. § I320c-9(a) (1982). 

11. Oster v. Bowen. 682 F. Supp. 853. 856-57 (E.D. Va. 1988); 28 U.S.C. § 2412(b) (1982 & Supp. IV 1986). 

12. 5 U.S.C. §8 551-559 (1982); but see J. Blum. P Gertman & J. Rabinow, PSROs and The Law 1 19-29 
(1977) (arguing that PSROs should be considered federal agencies under the APA and FOIA). 

13. See infra notes 98-99, 259-70. The question of whether PROs are federal actors has not come up in cases lo 
date. Most decided cases have challenged sanctions imposed by the QIC, which clearly is part of the federal governmeni. 
In these cases, the question of the constitutionality of PRO procedures arises only derivatively, as pan of the process of 
generaung sanctions ultinuuely imposed by the GIG. In one case, however. Kuown v. Southeast Mo. PSRO, 811 F 2d 
401 (8th Cir. 1987). cen. denied. 108 S. Ct. 1994 (1988). the court decided that PSROs were federal actors absolutely 
immune from civil righu claims under the Constitution and thus held unplicitly that they were in fact subject to the 
Constituoon. as the question of immunity would not arise were they not. 



80 TIMOTHY STOLZFUS JOST 

This Article examines both what the law with respect to PROs is and what the 
law should be. Pan 11 provides a general descnption of the PRO program. Parts III 
through X consider the following eight areas in which administration of the PRO 
program affects beneficianes. providers, and practitioners: the adoption by HHS of 
rules, policies, contracts, and instructions that govern the PROs; the formulation and 
dissemination of cnteria, norms, and standards by the PROs; the PRO process for 
sanctioning providers and practitioners who provide care that is unnecessary or of 
substandard quality; denial of payment for substandard care; the investigation by the 
PROs of beneficiary complaints regarding practitioners and providers; the PRO 
process for reviewing proposed hospital discharges of beneficiaries; the processes 
through which PROs deny payment for improper utilization of medical care and 
reconsider those denials; and the considerations governing PRO data dissemination 
and confidentiality. As to each of these subjects, this Article examines the current 
state of the law and practice and relevant policy considerations and recommends 
improvements in current procedures. 

This Article draws on a variety of information sources. It is based in part on a 
review of the relevant statute, regulations, manual instructions, and scopes of work 
governing the PRO program and the rapidly growing body of court cases and ALI 
opinions considering PRO issues. It also draws on dozens of articles from the medical 
and legal literature; testimony to congressional committees; position statements of 
affected groups; and reports from the General Accounting Office, HHS OIG, and 
other evaluators of the PRO program. Finally, it draws on interviews with nearly 
eighty informants, including seventeen representatives of national, state, and local 
beneficiary organizations; eleven representatives of provider groups; twelve attorneys 
who represent providers; twenty-three PRO representatives; and fourteen represen- 
tatives of the federal government. These included in-depth interviews with represen- 
tatives from twelve PROs (in most instances executive directors) reviewing the 
current procedures of those PROs in each of the areas under consideration. "* Finally, 
this Article considers comments received on an earlier draft from interested persons 
and organizations. 

II. The pro Program 

The initial Medicare law adopted in 1965 gave little attention to regulating the 
medical necessity, appropriateness, and quality of services provided Medicare 
beneficiaries. It required only hospital -based utilization review committees to assure 
appropriate utilization of services, '^ state licensure to assure that physicians were 
minimally qualified,'* and accreditation by the Joint Commission on Accreditation of 
Hospitals to guarantee the quality of hospitals. " By the early 1970s, however, it was 

14 The PRO repmentaiives inierviewed were from the suies of New Jersey. Pennsylvania. Virginia, Flonda. 
Illinois. Texas. North Dakota. Iowa. CaJifomia. Washington, Minnesou. and Michigan. The Iowa PRO also has the 
contract for Nebraska and the Washington PRO for Idaho and Alaska, so these programs were also discussed. 

15. Social Security Amendments of 1965. Pub. L. No 89-97. § 186l(k). 79 Stat. 285. 318-19 (1966). 

16. Id. § 186l(r). 79 Stat. 285. 321. 
17 Id. § 1865. 79 Stat. 285, 326-27. 



MEDICARE PEER REVIEW 81 

becoming apparent that further controls were needed to limit excessive utilization of 
Medicare services.'" Out of this concern grew the PSRO program, which used 
regional nonprofit physicians groups to review independently the use of medical 
services by beneficiaries of federal medical assistance programs, including 
Medicare." Though the primary emphasis of PSROs was on utilization review, they 
also conducted Medical Care Evaluation Studies (later Quality Review Studies) 
aimed at improving the quality of medical care. PSROs never succeeded in meeting 
the expectations of their supporters or overcoming the cnticisms of their increasingly 
vocal detractors. In 1982 the Tax Equity and Fiscal Responsibility Act (TEFRA) 
abolished the PSRO program and created in its stead the PRO program. ^^^ 

The PRO program was intended to be a leaner and more effective program than 
its predecessor. The 195 PSRO regions were trimmed to 54 statewide areas. The old 
system of grant-funding was replaced by biennial (now triennial) contracts, to be 
awarded by competitive bidding. Ineffective PROs were to be terminated. PROs 
could no longer delegate utilization review functions to hospitals, as had the PSROs. 
Though PROs were initially to be physician-sponsored organizations (as were the old 
PSROs), the statute allows HHS to turn to other organizations, including insurance 
companies or Medicare fiscal intermediaries, for PRO services if initial physician- 
sponsored contractors prove ineffective. Unlike PSROs, PROs could be for-profit 
entities. Finally, the PROs were given enhanced sanction and payment denial 
authority to enforce their power. 2' 

In the year following the creation of the PRO program (before it was in fact 
implemented). Congress adopted a prospective payment system (PPS) for Medicare 
based on diagnosis-related groups (DRGs)." This system ended the prior cost-related 
reimbursement system, under which Medicare had reimbursed hospitals the costs 
they incurred in caring for Medicare patients, and substituted for it a program that 
paid hospitals primarily on a lump sum per hospitalization basis. 

This change solved one problem addressed by the PROs, but created others. The 
old system had rewarded hospitals for keeping patients in the hospital as long as 
possible (which increased their costs and thus their reimbursement); thus a major 
focus of PSRO utilization review had been controlling the length of inpatient hospital 
stays. Because the new DRG system creates incentives for hospitals to minimize their 
costs by discharging patients as soon as possible, the problem of excessive 
length-of-slay has been solved. This problem has been replaced, however, by other 
serious problems: the DRG system creates incentives for hospitals to discharge 
patients prematurely, to underserve them while in the hospital, and to try to game the 



18. See Senate Comm. on Fin., S. Rep. No. 1230, 92d Cong.. 2d Sess. 254-69 (1972): Senate Comm. on Fin.. 
Medicare & Meoicaio, Problems. Issues and Alternatives. 91st Cong.. 2d Sess. 105-09 (Feb. 9. 1970). 

19. Social Security Amendments of 1972. Pub. L. No. 92-603. § 249F. 86 Stat. 1329, 1429-45 (1972) 

20. Tax Equity and Fiscal Responsibility Act of 1982. Pub. L. No. 97-248. § 143. 96 Stat. 324. 382 (1982). 

21. See CiSLOwsKi, Congressional Research Service Report for Congress. The Peer Review Orcanization 
Program 4-5 (1987); 1>ohr, Peer Review Organizations: QuALmr Assurance in Medicare (The Rand Paper Series) 7. 
14-15 (1985); Mellene. The Changing Focus of Peer Review Under Medicare. 20 U. Rich. L. Rev. 315 (1986) 
(describing the changes from the PSRO to PRO program). 

22. Social Security Amendments of 1983, Pub. L. No. 98-21. Title VI. 97 Stat. 65. 149-72 (1983). 



82 TIMOTHY STOLZFUS JOST 

system, eiiher by transfemng patients between hospitals or units within hospitals or 
by assigning improper DRGs. These practices can potentially lead to lower quality 
care for patients; therefore, every major budget reconciliation act since 1983 has 
included provisions redirecting the mission of PROs to address the potential quality 
and access problems created by the DRG reimbursement system. 

There are currently forty-four PROs serving the fifty-four PRO areas.-* 
Sixty-eight percent were formerly PSROs.-* Eighty-four percent are "physician- 
sF>onsored" — that is, they are either composed of at least 20% of the physicians 
practicing in the review area or of 10% of the physicians in the area and are otherwise 
representative of the state physician community." The remaining PROs are "phy- 
sician access" organizations, usually insurance companies, having a sufficient 
number of physicians available to carry on review functions.^* HHS is supp>osed to 
give preference in contracting to "physician-sponsored organizations," if any are 
available." PROs must include on their boards at least one consumer 
representative.^* The PROs vary significantly in size, the largest having hundreds of 
employees and budgets of millions of dollars. The proposed budget for the PRO 
program for fiscal year 1989 is $257.4 million, an increase of 50% over fiscal year 
1988. 

PROs are delegated review responsibility under contracts with HHS. Until this 
year, these contracts were bid on a two-year cycle, but now they last for three years.-' 
HHS may terminate a PRO that has substantially failed to carry out its contract.^" 
HHS monitors PRO performance primarily through three methods: (1) periodic data 
reporting from the PROs, (2) the PRO Monitoring Protocol and Tracking System 
(PROMPTS-2) regional office review system, and (3) the SuperPRO, an independent 
contractor, which verifies PRO reviews." The PRO program is supervised by the 
Health Standards and Quality Bureau (HSQB) of the Health Care Financing 
Administration (HCFA) of HHS. 

The primary tasks of the PROs are to process data concerning health care services 
provided to Medicare beneficiaries^^ and to intervene when these data indicate that 
services have been provided unnecessarily, inappropriately, or with inadequate qual- 
ity. Because hospitals consume over two-thirds of Medicare expenditures, PROs have 
focused their review traditionally on care provided to beneficiaries by doctors in 
hospitals. Recently many of the PROs have begun to review care provided by health 



IZ. GIG £JTECTivF.,£5s Retokt. supra note 9. at 3. 

24 Id. 

25. Id.. 42 C.F.R. } 462.102 (1987). 

26 42 C.F.R. 5 462.103(1987) 

27 42 U.S.C. § l320c-2(bKI)(1982). 
28. Id. § 1320c- 1(3) (Supp. IV 1986). 

29 Id. § I320c-2(CM3)(I982). 

30 Id. § I320c-2(c)(6). Such a lennination is not subject to judicial review. In re Pa. Peer Review Og.. SO Bankr. 
640 (1985); 42 U.S.C. I 1320c-2(f) (1982). 

3 1 GIG EfFEcnvENESS Rekmt. supra note 9. at 12-18; GAG. Medicaiie: Imhiovino (}uauty of Caxe Assessment 
Kto As$UKA.NCE 53-58 (May 1988). 

32. PROs also review Medicaid cases for slates that contract for such assistance, 42 U.S.C. i 1320c-7, aiMl review 
for private payers, id. i 1320c-3(a)<ll). 



MEDICARE PEER REVIEW 83 



maintenance organizations and competitive medical plans (HMOsyCMPs) with 
Medicare risk-shanng contracts." The Omnibus Budget Reconciliation Act of 1986^ 
(OBRA '86) also requires PROs to stretch their review capacity to cover services 
provided in other settings, including post-acute care provided by skilled nursing 
facilities and home health agencies; ambulatory and hospital outpatient care; and 
beginning in 1989 care provided by physicians in their offices.^* Most PROs, 
however, still focus the vast majority of their resources on review of care provided 
in hospitals, the primary concern of this Article. 

The principal source of data for PRO review is the hospital record. PROs 
regularly receive from fiscal intermediaries (the insurance companies and other 
entities that handle Medicare reimbursement to providers) data on bills paid for 
services rendered to Medicare beneficiaries. The PRO selects a sample of these cases 
for review and requests medical records on these cases from the hospitals, which are 
reviewed at the hospital or at the PRO office.^* The sampling criteria that PROs use 
for selecting cases for review, and the focus of their review in examining the records, 
have varied over the three contract cycles during which PROs have been in operation. 
During each contract cycle, the screening criteria and focus of PRO activity have 
been established by a scope of work.'' 

The sampling criteria mandated by the Third Scope of Work, currently being 
implemented, require a PRO to review, for each PPS hospital (hospitals reimbursed 
under Uie DRG prospective payment system) under its jurisdiction, a 3% random 
sample of all discharges; 50% of cases involving transfers from one PPS hospital to 
another; 10% of transfers to a psychiatric bed in a PPS hospital (and 100% of certain 
problem transfers to psychiatric beds); 25% of transfers from a PPS hospital bed to 
a nursing home bed in the same hospital; 25% of cases in which a patient discharged 
from a PPS hospital is readmitted within thirty -one days; 20% of cases in the 25% 
discharge and readmission sample Just mentioned, in which the patient received care 
from a nursing home, home health agency, or hospital outpatient area during the 
period intervening between hospitalizations; 25%, 50%, or 1(X)% of cases coded with 
certain problem DRGs; 25% of day and cost outliers (cases in which hospitals 
received extra payment beyond the DRG reimbursement because the case required an 
extraordinarily long or expensive hospital stay); all cases with targeted principal 
diagnoses, such as obesity or pacemaker fitting or adjustment; all cases in which a 
hospital has requested that a case be adjusted from a lower to a higher DRG; all cases 
in which a hospital has determined that an admission was not covered but the patient 
required Medicare-covered care at some time during the stay; and all cases referred 



33 5rf Consolidated Omnibus Budget Reconciliation Act of 1985. Pub. L. No. 99-272. § 9405. 1986 U.S. Code 
Cong. & Admin. News (100 Stat.) 83. 201 (hereinafter COBRA '85). as (vnrm/f(y<>y Omnibus Budget Reconciliation Act 
of 1986. Pub. L. No 99-509. § 9353(a). 1986 U.S. Code Cong. & Admin News (100 Stat.) 1874. 2044 |hereinafter 
OBRA •86). 

34. OBRA 86. Pub. L. No. 99-509. 1986 U.S. Cooe Cong. & Admin. News (100 Stat.) 1874. 

35. Id. § 9353(a). (f). 1986 U.S. Code Cong. & Admin News (100 Sui.) 1874. 2046. 2052. 

36. A recent study found dial 61% of PROs usually peifonn review on-site at the hospital; 20% always perfonn 
it on-site. OIG Effectiveness Retokt. supra note 9, at 3. 

37. For a funhcr description of a scope of work, sec infra text following note 56. 



84 TIMOTHY STOLZFUS JOST 



to the PRO by the fiscal intermediary or by HCFA.^* PROs are also required to 
review a random sample of 15% of discharges from non-PPS units of PPS hospitals 
(for example, rehabilitation units) and from non-PPS-reimbursed hospitals and a 5% 
random sample of cases from ambulatory surgical centers.^' When reviews indicate 
that a hospital is committing errors in more than 5% of its cases (or six cases if this 
amount is greater), the PRO is to intensify review to 50% or 100%, depending upon 
the problem, of the hospital's Medicare cases. •*** 

To this point, all the reviews listed are retrospective. The PROs must also 
f>erform preadmission or preprocedure review of ten specific procedures and the use 
of assistants for cataract surgery."' Finally, a separate HMO/CMP Scope of Work 
provides a sampling procedure for identifying HMO/CMP cases to be reviewed. The 
intensity of sampling of HMO/CMP cases is related to the confidence that the PRO 
has in the HMO/CMP's own internal quality control capacity. In total, sample cases 
under the second contract cycle totaled about 26% of all Medicare hospital 
admissions.*^ 

Once medical records fitting these sample criteria are identified and copied, they 
are reviewed by professional reviewers (usually nurses), who apply criteria screens to 
identify utilization or quality problems. This review must be completed within sixty 
days from the date of receipt of the list of cases from the fiscal intermediary.*^ Each 
inpatient hospital discharge is to be reviewed for quality using HCFA's generic 
quality screens, for necessity and appropriateness using PRO discharge and admis- 
sion criteria screens, and for DRG validation.** Care provided outside of the hospital 
setting is reviewed only for quality and not for utilization problems. 

Once a PRO identifies a problem through this review of medical records, the 
case is routed to a physician reviewer. If the physician confirms the problem, the case 
can go in one of two directions. First, if a quality problem is identified, the case is 
routed to the PRO quality assurance system, which can interpose various interven- 
tions (including sanctions ultimately) to correct the problem.*^ If, on the other hand, 
the problem is identified as a utilization problem, the case is considered for a payment 
denial.** PROs also continually assemble profile data in an effort to identify aberrant 
providers and physicians. Profiles are kept on patients, physicians, hospitals, DRGs, 
diagnoses, and procedures to monitor PRO impact and identify problems for further 
study.*' 

PROs have a number of other functions unrelated to their data gathering and 
analysis functions. They are responsible for reviewing cases when hospitals inform 



38. Third Scope of Work, supra note 3, § VI. at 16-22. 

39. Id. §§ Vll-Vlll. at 22-23. 

40. Id. § IX. ai 24-26. 

41 Id. § X. at 26-27. 

42 HCFA Official Ouilines PRO Program Objectives. HosPfTALS. Aug. 5. 1987. at 122. 

43. Third Scope of Work, supra note 3. § III(B). at 2-3. 

44. These screens, and problems associated with them, are descnbed further in Pan IV of this Article. 

45. The PRO quality assurance/sanction process is descnbed in Pan V of this Article. 

46. Payment denials and reconsiderations are considered in Part IX below. 

47. Third Scope of Work, supra note 3. § XV, at 29-32. 



MEDICARE PEER REVIEW 85 



patients that their care is not, or is no longer, covered by Medicare.-^ PROs are also 
responsible for monitonng to assure that hospitals provide beneficiaries with a 
statement of their nghts to PRO discharge review at the time of admission/' They are 
required to investigate complaints by Medicare beneficiaries about the quality of 
Medicare-covered services received from Medicare-certified hospitals, nursing 
homes, home health agencies, or ambulatory surgical centers.^ Finally, PROs are 
resp>onsible for educating beneficiaries and providers as to their existence and 
functions.*' 

III. Publication of PRO Program Policies and Procedures 

One major administrative law issue that has arisen under the PRO program 
concerns the extent to which directives and guidelines governing the program must 
be promulgated as rules under the APA. At the time the program was initiated in 
1982, some regulations survived from the PSRO program that preceded it. Additional 
regulations have been promulgated that address issues such as PRO eligibility, area 
designations, imposition of sanctions, confidentiality and disclosure, reconsidera- 
tions and appeals, and review activities. '^ 

These regulations, however, represent only a small portion of the instructions 
HHS has issued to govern the PRO program. First, HHS has issued a PRO Manual, 
supplemented fjeriodically by manual transmittals and interim manual instructions. 
Chapters in the Manual address such issues as PRO review procedures; PRO denials, 
reconsiderations, and appeals; waiver of liability determinations; sanctions; data and 
reports; and PRO administration. Some portions of the Manual track closely the 
statute and regulations.'' Even these sections, however, cover details not addressed 
by the regulations.^ Other parts of the Manual cover issues not addressed by the 



4«. 42 use. § I320c-3(e) (Supp. IV 1986). This procedure is discussed at Pan VIII below. 

49. Third Scdpe of Work, supra note 3, § XVI. at 33-34. 

50. 42 use. § 1320c-3(aM14) (Supp. IV 1986); Third Scope of Work, supra note 3. § XVII(A). at 36-38 

51. 42U.SC. § l320c-3(aK4)(B)(Supp IV 1986); Third Score of Work, jupra note 3. § XVI(C). at 34-35 For 
funher background on the PRO program, refer to Cislowski. supra note 21; Ix)hr. supra note 21. Mellette, supra note 
21. 

52. The initial set of regulations was promulgated at 401 Fed. Reg. 7.201 (Feb. 27. 1984) and 50 Fed. Reg. 15.311 
(Apr. 17. 1985) and codiHcd at 42 C F R. pts. 400. 405. 420. 422. 431, 433. 456. 460. 462, 466. 473. 474, 476, and 
489 (1987). On September 30. 1986, rules governing imposition of sanctions were redesignated to 42 C.F.R. pt. 1004 
(1987). On October 7. 1987. HHS published rales govermng PRO HMO/CMP review. 52 Fed. Reg. 37,545 (Oct 7. 
1987). These rales were published as final without a notice of proposed ralemaking (NPRM) under the APA good-cause 
excrption to efiectuate iii a "timelv manner" (provision: of OBRA '86, adopted in October of 1986, which were to be 
in effect by Apnl of 1987 52 Fed. Reg. 37,456 (Oct. 7, 1987). Corrections to this notice were published in December 
of 1987, 52 Fed. Reg 47.003 (Dec. II, 1987). Finally, on March 16, 1988, HHS proposed further regulations to 
implement provisions of COBRA '85. Pub L. No. 99-272. 1986 US. Code Cong. & Admin News ( 100 Stat ) 83. and 
OBRA "86. Pub L. No. 99-509, 1986 U.S. Code Cong & Admin. News (100 Stat.) 1874, and to make technical changes 
necessitated by experience with the PRO program. 53 Fed. Reg. 8.654 (Mar. 16. 1988). Additional provisions rcquinng 
PRO review of proposed changes in DRG assignments were also published at 52 Fed. Reg. 33.033. 33,059-60 (Sept. 
1. 1987). 

53. See. eg. U.S. De^'t of Health & Human Servs.. Peer Review Organization Manual §§ 4080-4086. 
Reconsiderations and Appeals (1988) (hereinafter PRO Manual). 

54 53 Fed. Reg 8.666 (Mar. 16, 1988). For example, PRO Manual § 4084 requires that a reconsideration 
reviewer be a physician who practices in a similar setting to that of the physician being reviewed whenever possible and 
thai the physician be board-cenined or board-eligible in the specialty of the reviewed physician. 42 C.F.R. § 473.28 



86 TIMOTHY STOLZFUS JOST 



regulations but clearly internal to the PROs. such as the data and reporting 
requirements appearing in detail in Manual Chapter 8. Finally, some Manual 
provisions that have effects external to the PROs differ matenally from the PRO 
regulations." PRO activities are also affected by provisions in other HCFA manuals, 
such as the Medicare Hospital Manual.^'' 

The other central documents governing the PRO program are the PRO scopes of 
work and PRO contracts. HHS has recently issued the Third Scope of Work for the 
third contract cycle, and last year it issued a separate Scope of Work for review of 
HMOs and CMPs. HHS is currently entering into contracts reflecting the Third Scope 
of Work. The Scope of Work specifies in great detail PRO review responsibilities and 
data requirements and incorporates by reference the PRO statute, regulations, and 
PRO Manual. PRO contracts specify in even greater detail the review responsibilities 
and specific objectives of individual PROs. Finally, PROs are also governed by a 
variety of less formal instructions, such as regional medical review letters. 

The practice of relying on manual transmittals, program instructions, and con- 
tracts in lieu of regulations promulgated through notice and comment rulemaking is 
not unique to the PRO program. Throughout its administration of the Medicare and 
Medicaid programs, HHS has often used such matenals in preference to rules adopted 
through the APA notice and comment process." It is not difficult to understand HHS's 
preference for issuing instructions informally. Public notice and comment rulemaking 
has always been burdensome. The requirements of 5 U.S.C. section 553 that the public 
be given an opportunity to comment on proposed rules, that the agency consider the 
comments, and that publication of final rules precede their effective date by at least 
thirty days necessarily slow down the process of implementing policy. Recent re- 
strictions on notice and comment rulemaking, including Executive Orders No. 12,291 
and No. 12,498, requiring Office of Management and Budget (0MB) review of some 
rules, and 5 U.S.C. section 603, requiring regulatory flexibility analyses, have made 
APA rulemaicing increasingly burdensome and time-consuming.** It can easily take 
a year or more to promulgate a rule under these procedures.'' Regulations imple- 
menting sections 9401 and 9403 of COBRA '85<* have been under consideration 
for over two years and have only very recently been published as a notice of 

currently only requires ihal the leconsideralion physician be a specialist in ihe type of services under review, though 
proposed rule 473.28 (proposed on March 16, 1988) is identical lo the PRO Manual provision. 

55 PRO Manual § 6025. for exainple. implements the 1987 HCFA/AMA/AARP compromise on sanction 
procedures, di&cussed iit Part V b»lGw .■•pd pmvid's pnceui-ril protections to providers and practitioners beyond thoce 
specified ui 42 C.F R. p«. 1004. 

56. See HtM.THCAKE Fin. Aomin.. U.S. Def'tof Health & Human StRvs.. Medicare Hosf. Manual § 312 (July 
1988) (dealing with notices to beneHcianes of PRO review of care) (hereinafter Medicare Hosp Manual). 

57 See St. Marys Hosp. of Troy v. Blue Cross Blue Shield. 788 F 2d 888 (2d Cir 1986): Cubanski v. Heckler. 
781 F 2d 1421 (9th Cir. 1986); Powderly v Schweiker. 704 F 2d 1092 (9th CIr 1983). Rio Hondo Memonal Hosp v 
United States. 689 F.2d 1025 (Cl. CI. 1982); Samuels v Heckler, 668 F Supp. 656. 664 (W D Tenn 1986): New York 
SuieOep't of Social Serv. v Bowen. 64« F Supp 850(D DC. 1986): Bond Hosp. Inc v Heckler, 587 F Supp 1268, 
1272 (DOC 1984), affd. 762 F 2d 137 (DC Cir. 1985): Kinney. The Medicare Appeals System for Coverage and 
Paymeni Disputes: Achieving Fairness in a Time of Constraint. I Admin. L.J. 54 (1987) 

58. Although E«ec Order No. 12,291, 46 Fed Reg. 13,193. only requires OMB review of "major rules," in fact 
CMS reviews all rules proposed by HCFA 

59. DeMuth A Ginsberg. White House Review of Agency Rulemaking. 99 Harv L Rev 1075. 1 159 (1986). 

60. COBRA -85. Pub. L. No. 99-272. 1986 U.S. Code Cong. & Admin. News (100 Stat.) 83, 196-200. 



MEDICARE PEER REVIEW 87 

proposed rulemaking (NPRM).*' Given the rapid changes in the Medicare and 
Medicaid programs generally and in the PRO program in particular, it is not 
surpnsing that HHS has sought means other than notice and comment rulemaking for 
program management. 

While HHS's eschewal of notice and comment rulemaking may be understand- 
able, it is not necessarily right. There are sound policy reasons grounding the APA's 
requirement of notice and comment rulemaking. First, it "reintroduces public 
participation and fairness to affected parties after governmental authority has been 
delegated to unrepresentative agencies."" It also assures a prepublication dialogue, 
which permits persons affected by a rule to educate the agency as to their concerns. 
This in turn allows the agency to achieve a more rational regulatory scheme or a more 
intelligible explanation of the scheme it originally prof)osed." It forces the agency to 
think carefully about its policies, so as to justify them before a skeptical public.** It 
"assures that the agency will have before it the facts and information relevant to a 
particular administrative problem, as well as suggestions for alternative solutions."" 
Finally, public participation in rulemaking contributes to public acceptance of the 
legitimacy of the regulatory result." The Administrative Conference of the United 
Slates (ACUS) has consistently urged HHS to provide greater opportunity for public 
comment on its policies for making coverage and payment determinations.*"^ 

More specifically, the attempts of HHS to govern the PRO program without 
using the APA rulemaking process have subjected the PROs to a continual and 
confusing stream of instructions, which have severely hampered their ability to carry 
out their mandate.** They have also angered the providers and practitioners governed 
by the program. 

In October 1984 an association of those providers, the American Hospital As- 
sociation (AHA), filed a petition with HHS for rulemaking,** which requested HHS 
to promulgate comprehensive regulations for the PRO program. When HHS failed to 
do so, the AHA brought suit claiming that HHS had violated the APA. The District 
Court for the District of Columbia held that HHS had indeed violated the APA by 
promulgating the PRO Scope of Work, contracts, and several manual transmittals 
without notice and comment rulemaking. ^o HHS appealed this determination to the 



61 . Medicare and Medicaid Programs; Denial of Payment for Substandard Quality Care and Review of BeneHciary 
Complaints. 54 Fed. Reg. 1956 (1989) (to be codined at vahous parts of 42 C.F.R.) (proposed Jan. 18. 1989) (proposed 
rule) Iheremafter PROfOSO) Substandard Cajie Reculation''| See also infra Part VI. 

62. Banerton v. Marshall, 64« F 2d 694. 703 (DC. Cir. 1980); see also National Assn of Home Health Agencies 
V. Schweiker. 690 F.2d 932. 949 (DC. Cir. 1982). cert dented. 459 U.S. 1205 (1983). 

63. American Bus Assn v. United Slates. 627 F 2d 525. 533 (D.C. Cir. 1980). 

64. New Jersey v Department of Health &. Human Servs.. 670 F.2d 1262 (3rd Cir. 1981). 

65. Guardian Fed. Sav & Loan v. Federal Sav & l^oan Ins. Corp.. 589 F.2d 658. 662 (DC Cir. 1978). 

66. Chamber of Commerce of the United Sutes v. OSHA. 636 F.2d 464 (DC. Cir. 1980) 

67. 1 C.F.R. §§ 305.86-5. 87-8 (1987). 

68. Baldwin & Fackelmann. Blizzard of Paperwork. New Rules are Burying PROs and Hospitals. Mou. 
Healthcare. Jan. 3. 1986. at 46. 47-48. 

69 See 5 U.S.C § 553(e) ( 1982). 

70. American Hosp. Assn v. Bowen. 640 F. Supp. 453 (D.D.C. 1986). The court also held two manual 
transmittals to be valid interpretive rules, not subject to notice and comment rulemaking. This holding was not appealed. 



88 TIMOTHY STOLZFUS JOST 

District of Columbia Circuit Court of Appeals, which reversed the distnct court. ' ' The 
appellate court's majority opinion, written by Judge Wald, found the contract. Scope 
of Work, and manual transmittals to have been covered by exceptions to the APA. 
Judge Mikva dissented in part, arguing that the challenged contract objectives should 
have been promulgated through APA rulemaking procedures.'^ Ultimately, Congress 
seems to have had the last word, as provisions of the Omnibus Budget Reconciliation 
Act of 1987 (OBRA '87)^' call for publication of PRO contract provisions, though not 
necessarily using APA procedures. Before turning to this legislation, the rulemaking 
requirements of the APA in general, and the interpretation of them in American 
Hospital Association v. Bowen in particular, will be considered in greater detail. 

The notice and comment and publication requirements of 5 U.S.C. section 553 
are subject to a number of exceptions, several of which arguably apply to the PRO 
program. Most obviously, section 553 does not apply to "a matter relating to . . . 
public property, loans, grants, benefits or contracts."'* As PROs assist in running a 
benefit program (Medicare) under contract with HHS, instructions governing the 
adnunistration of the PRO program would seem to fall within this exception. This 
exception has been widely criticized as an atavistic survival of simpler days when 
government benefit programs were uncommon rather than pervasive and when the 
law still drew a clear line between rights and privileges. '' The Secretary of HHS has 
yielded to these criticisms and voluntarily waived the protection of this exception.'* 
HHS is legally bound by this waiver. '^ 

In its brief in American Hospital, HHS argued that, even though it has generally 
waived the APA contract exception, the PRO statute expressly exempts PRO 
contracts from any APA constraints.'* 42 U.S.C. section 1320c-2(e) provides that the 
Secretary's authority in making PRO contracts is not to be trammeled by "any 
provision of law relating to the making, performance, amendment, or modification of 
contracts of the United States."" This provision was adopted by Congress to 
promote flexibility and avoid restriction of "innovation in new approaches to 
review. "80 The D.C. Circuit, however, rejected this argument, noting that the 
provision exempted PRO contracting from "the vast corpus of law establishing rules 
regarding the procurement of contracts from the government"*' and not from the 



71. American Hosp. Awn v. Bowcn, 834 F.2d 1037 (DC. Cir. 1987). 

72. Id. ax 1058-62 (Mikva. J.. dUseming). 

73. Omnibus Budget Reconciliation Act of 1987. Pub. L. No. IOO-2C3. 1987 U.S. Code Cong. <Sc Admin. News 
(101 Sut.) 1330 Ihereinafter OBRA '87|. 

74. 5 US.C. § 553 (1982). 

75. See BonHeld. Public Participaiion in Federal Rulemaking Relating to Public Property, Loans. GrantM, 
Benefits, or Contracts. 1 18 U. Pa. L. Rev, 540 (1970); 1 C.F.R. § 305.69-8 (1987). 

76. 36 Fed. Reg. 2.532 (Feb. 5, 1971) 

77. Herron v. Heckler. 576 F. Supp. 218. 229 (N D. Calif. 1983). 

78. Brief for Appellani al 35-39. American Hosp. Ass'n v. Bowen. 834 F.2d 1037 (DC. Cir. 1987) (No. 
86-5579). 

79. 42 use. i I320c-2(e)(l982). 

80. S. Ret. No. 494. 97ih Cong.. 2d Sess. 41. 43. reprinted in 1982 U.S. Code Cong. & Aomin. News 817-19. 

81. American Hosp.. 834 F.2d al 1054. 



MEDICARE PEER REVIEW 89 



APA The court noted that nothing in the provision's legislative history indicated that 
Congress intended to retract HHS's own waiver of the APA contract exemption ""^ 

Because HHS was precluded by its waiver from relying on the grants, benefits, 
and contracts exception in American Hospital, it instead relied primarily on 
exceptions found in 5 U.S.C. section 553(b)(3)(A) for "interpretive rules, general 
statements of policy, or rules of agency organization, procedure, or practice.""' The 
law governing these exceptions to the APA is tembly obscure, descnbed by various 
cases as "tenuous." "blurred." "baffling," and "enshrouded in considerable 
smog."** 

In the district court. HHS argued unsuccessfully that the manual instructions 
were interpretive rules. Interpretive rules are provisions that "merely clarify or 
explain existing law or regulations."" They track and fine-tune statutory or 
regulatory requirements or remind regulated individuals or entities of existing 
duties.'*'' elucidating what an administrative officer thinks a statute or rule means. "^^ 
They have no independent force of Iaw.*» 

Some provisions of the PRO Manual, scopes of work, and contracts merely 
restate and clarify statutory obligations. Examples include provisions dealing with 
data confidentiality or disclosure. But most provisions of the PRO Manual and 
contracts address issues not directly covered by statutes and regulations, such as 
review sampling and data reponing requirements, and thus cannot properly be 
classified as interpretive rules. The district court in American Hospital so held,*'' and 
the court of appeals affirmed.'** 

The court of appeals, however, relied on other exceptions found in 5 U.S.C. 
section 553 \p uphold the PRO Manual instructions, contract, and Scope of Work. 
First, it held that several manual instructions focusing PRO review on particular 
objectives were validly exempt from notice and comment rulemaking because they 
were procedural rules, exempt under 5 U.S.C. section 553(b)(A).9' The procedural 
rule exception exists to allow agencies flexibility in arranging their internal 
operations — "it covers agency actions that do not themselves alter the rights or 
interests of panies, although it may alter the manner in which parties present 
themselves or their viewpoints to the agency. "'^ As procedural rules do not directly 



82. Id. 

83 5 U S.C. § 553(bK3MA) (1982). 

84. See. f.j;.. Community Nutntion Inst. v. Young. 818 F. 2d 943. 946 (D.C. Cir. 1987) (and cases cited therein) 
S5. Alc.-raz v. Block. 746 r.2d 593. 613 (9th Cir. :084) (quoung Powdwly v Schweikcr. 704 F.2d 1092. ;098 
(9th Cir. 1983)). 

86 Cabais v. Egger. 690 F.2d 234. 238 (DC. Cir. 1982). See A&simow. Nonlegislative Rulemaking and 
Regulatory Reform. 1985 DuKE L.J. 381. 393-97. 

87 Amencan Postal Workeis Union v United Sutes Postal Serv.. 707 F 2d 548 (DC. Cir. 1983). cen denied. 
465 US. 1100(1984); In Home Healthcare V. Bowen. 639 F. Supp. 1124 (D DC. 1986) 

8«. HciTon V. Heckler. 576 F Supp. 218. 231 (N.D. Cai. 1983). 

89 Amencan Hosp. Assn v Bowen. 640 F. Supp. 453. 462-63 (D.D.C. 1986). 

90 Amencan Hosp. Assn v. Bowen. 834 F.2d 1037, 1050 (D.C. Cir. 1987). 

91 Id. at 1049-51. 

92. Banenon v Marshall. 648 F.2d 694. 707 (DC. Cir. 1980) (rootnote omitted); see also Neighborhood TV Co 
V. FCC. 742 F.2d 629 (DC. Cir. 1984); Mayton. A Concept of a Rule and the Substaiuial Impact ' Test in Rulemaking. 
33 Emmy L.J. 889. 900 (1984). 



90 TIMOTHY STOLZFUS JOST 



govern public conduct, it is argued, their formation does not require the public 
participation mandated by the APA.**^ In American Hospital, Judge Wald articulated 
the key test for identifying a procedural rule as "whether the agency action also 
encodes a substantive value judgment or puts a stamp of approval or disapproval on 
a given type of behavior.'"** A rule governing procedure that does not enforce such 
a substantive value judgment is exempt from APA rulemaking requirements, 
regardless of its actual impact on the nghts of those affected by agency action. This 
interpretation expands the procedural rule exemption and is at variance with earlier 
authorities, which placed emphasis on whether the rule in fact had a substantial 
impact on substantive rights. '' 

Applying this definition of procedural rule. Judge Wald upheld as exempt from 
APA rulemaking requirements manual provisions establishing sampling procedures 
for targeting PRO review. This holding is consistent with other cases holding that 
strategies for enforcement or timing of review are procedural in nature.'* It is, 
however, based on two questionable premises. 

First, Judge Wald's opinion is clearly based on the proposition that PROs are, 
in effect, agents or extensions of HHS. If the PROs are in reality a part of HHS, the 
disputed manual transmittals establishing procedures to be used by the PROs in effect 
dictate the internal procedures of a federal agency. '^ Following this line of reasoning. 
Judge Wald posited that any impact of the challenged HHS directives on PROs was 
not relevant to the question of whether such directives had a sufficiently substantial 
effect to render them substantive rather than procedural rules. 

PROs arc, of course, federal entities for some purposes.'* But they are also 
independent, private corporations, contracting to provide a service to the 
government." If PROs are part of the federal government solely because they 
provide a service under contract, so are Medicare carriers and intermediaries and, for 
that matter, hospitals and physicians who provide services as agents of the 
government to the ultimate beneficiaries of the Medicare program. In fact, when 
HHS has engaged in notice and comment rulemaking, it has considered the impact of 
its rules on PROs as if they were "small entities" under the Regulatory Flexibility 
Act, thus recognizing their independent existence.'*'*' 

It could be argued, of course, that the federal government should be able to deal 
with its contractors without being bothered by notice and comment rulemaking, and 



93. United Stau:5 Dep't of Labor v Kast Metals Co.. 744 F.2d 114S (5th Cir. I9ft4). 

94. American Hosp . 834 F 2d at 1047. 

95. See. e.g.. Pickus v. United States Bd. of Parole. 507 F 2d 1 107. 1 1 12-13 (DC. Cir. 1974). 

96. See International Union. UAW v. Brock. 783 F 2d 237. 251 n. 18 (DC. Cir. 1986); Kasi Metals, 744 F.2d at 
WAS. Neighborhood J\. 742 F 2d at 629: Donovan v. Wollasion Alloys Inc.. 695 F 2d 1.9(tstCir. 1983); Davis Walker 
Corp. V Blumenthal. 460 F. Supp. 283 (D.D.C. 1978). 

97. American Hosp.. 834 F 2d at 1048-49. 

98. See Kwoun v. Southeast Mo. PSRO. 81 1 F.2d 401 . 407 (8di Cir. 1987). cert, denied. 108 S. Cl. 1994 ( 1988) 
(holding PSRO ofTicials to be federal officers for application of immunity doctrine in damage action); Smith v. North La. 
Medical Review Assn. 735 F 2d 168. 173 (5th Cir. 1984) (holding PSRO to be federal rather than state entity for 
application of state action requirement). 

99. See supra notes 7-9 and accompanying text. 
100. See 53 Fed. Reg. 8.662 (Mar. 16. 1988). 



MEDICARE PEER REVIEW 91 

as noted earlier, the APA explicitly exempts from notice and comment rulemaking 
matters pertaining to government contracts. Yet, HHS has waived the protection of 
this exemption generally, and this waiver includes within its scope, presumably, 
HHS's governance of PROs as well as of hospitals. Turning again to the policies 
underlying the APA notice and comment requirement, PROs indisputably have an 
interest in participating in the process used for determining their responsibilities and 
have valuable information to contribute to that process. Thus, Judge Wald's 
conclusion that the substantive impact of HHS rules on PROs is irrelevant to a 
consideration of whether such rules are procedural or not is simplistic and probably 
wrong. The AHA may not have had standing to challenge the effect of HHS's 
informal policy on the PROs, but this issue is a real one. 

Second, Judge Wald's opinion is also based on the notion that manual 
transmittals directing enforcement strategy have only a minimal impact on hospitals 
and other providers. Providers, of course, have no legitiniate interest in freedom from 
effective oversight, but they may well have an interest in avoiding enforcement 
strategies that create excessive burdens '"' or that unfairly single out particular classes 
of providers. '<'2 Other courts have noted that enforcement strategies may have 
substantive effects. '°^ 

HHS did not take the position that its enforcement strategy was secret; indeed, 
HHS has described this strategy in the PRO Manual. Having chosen to make its 
enforcement strategy public, HCFA would not have suffered a great additional 
burden in seeking information from the public as to the wisdom of that strategy. 

The court of appeals upheld the validity of the PRO Scope of Work and contracts 
under another exception to section 553, the "general statement of policy" exception. 
This exception to the APA is probably the most mysterious. Cases attempting to 
explicate it describe "general statements of policy" as akin to press releases, 
announcing to the public an agency's intention as to what plans and priorities it will 
seek to establish in the future. '»* Policy statements do not create binding norms that 
constrain the agency's discretion."^' They are tentative and do not foreclose agency 
alternatives or conclusively affect rights.'"* Some cases exclude from this category 
statements having a substantial impact on substantive rights,'"^ while others reject 
this distinction, 'o* Most cases have considered the binding nature of a pronouncement 



101 . Cf. Burlington McmonaJ Hosp. v. Bowen. 644 F. Supp. '0?0 (WD. Wis. 1986) (considenng HCFAs alleged 
rrspunsib'lity to rr!r.ibi;rse providers for costs of phoiocopyirg necessitated oy PRO review). 

102. See infra text accompanying notes 169-70. 

103. Community Nutrition Inst. v. Young. 818 F.2d 943 (D.C. Cir. 1986); Estate of Smith v. Bowen. 656 F. Supp. 
1093 (D. Colo. 1987). 

104. Mada-Luna v. Fitzpatnck. 813 F 2d 1006. 1013 (9th Cir. 1987); Pacific Gas & Elec. Co. v. Federal Power 
Comm n. 506 F.2d 33. 38 (DC. Cir. 1974). 

105. Mada-Luna. 813 F.2d at 1014. 

106. Battcrton v. Marshall. 648 F 2d 694. 702 (DC. Cir. 1980). 

107 W C. V Bowen. 807 F.2d 1502 (8th Cir. 1987); Pickus v. United States Bd. of Parole. 507 F.2d 1 107. 1112 
(DC. Cir. 1974). 

108. Jean v. Nelson. 71 1 F.2d 1455 (1 1th Cir. 1982). aff'd. 472 U.S. 846 (1985). See also Note. An Analysis of 
the General Statement of Policy Exception to Notice and Comment Procedures, 73 Geo. L.J. 1007 (1985) (arguing for 
both a binding nonn and substantial effect test). 



92 TIMOTHY STOLZFUS JOST 

to be decisive in determining whether it is a "general statement of policy" or not.'"^ 
Policy statements that narrow the field of vision of a decisionmaker, minimizing the 
influence of some factors and encouraging decisive reliance on other factors that 
might not have been decisive had rulemaking procedures been followed, must be 
promulgated as rules. "^ 

Judge Wald's position that PRO scopes of work are general statements of p>olicy 
is certainly plausible. They are. after all, pnmanly statements by HHS of what it 
hopes to achieve through its contracts. The court's holding that PRO contracts are 
mere general statements of policy, however, is more questionable. In particular, the 
court's argument that objectives found in the contracts are not binding upon the PROs 
and thus will have no substantial effect on the PRO application of standards in 
necessity and quality review seems highly suspect. One of the primary arguments 
relied on by Congress for replacing the PSRO program with the PRO program was 
that the PROs would be more effective in controlling utilization and quality precisely 
because they would be held to their contractual objectives under threat of 
nonrenewal.'" In fact, some PRO contracts were not renewed after the first contract 
cycle because the PROs failed to meet contract objectives, "^ and HCFA's evaluation 
of PROs for renewal of contracts in the second cycle relies heavily on the PROs' 
contractually assumed review activities."^ Indeed, when HHS has engaged in 
rulemaking, it has acknowledged the significant effects that review of PROs pursuant 
to their contracts will have on providers, practitioners, and beneficiaries."* It is 
disingenuous for HHS to say, therefore, that PRO contracts have no binding effect on 
the PROs and thus no impact on beneficiaries or providers. 

Judge Mikva, in dissent, pointed persuasively to the analogy between the PRO 
contract objectives and efforts of the Social Security Administration to target its 
review on ALJs who granted a disproportionate number of disability awards. This 
effort was held earlier by the Ninth Circuit to be subject to APA rulemaking 
requirements because it effectively discouraged disability awards."' Similarly, PRO 
contract objectives undoubtedly have an impact on PRO payment denials, which in 
turn affect provider and physician behavior and, ultimately, the services received by 
Medicare beneficiaries. 

All of this is not to say that Judge Wald's opinion reaches the wrong conclusion. 
If its legal foundations are questionable, its result is certainly politic. In creating the 
PRO program. Congress intended to establish a system that could respond rapidly, 
flexibly, and creatively to developments in the delivery of health services. To impose 



109 Sre Community Nutrition Inst. v. Young, 818 F 2d 943. 946-48 (DC. Cir 1987); Telecommunications 
Research & Action Center V. FCC. 800 F 2d 1181. 1186 (DC. Cir. 1986): Brock v. Cathedral Bluffs Shale Oil Co.. 7% 
F 2d 53.^ (DC. Cir. 1986) (use of permissive or mandatory language may be decisive): Amencan Bus Ass'n v. United 
Slates. 627 F 2d 525. 529 (DC. Cir. 1980). 

no Pickus. 507 F2d at 1113 

111 SeeS Rep. No. 494. 97th Cong.. 2d Sess. 41, repnmed m 1982 US. Coot Cone. & Admin News 781. 817: 
42 use § I320c-2(CK7)(I982). 

1 12. Growing Conlracl Dentals Dispirit ihe Nation's PROs. Hospttals. May 20. 1986. at 28. 

1 1 3 OIG Ejtectiveness Rekxjt, supra note 9. at 2 1 . 

114 50 Fed. Reg. 15.322-25 (Apr. 17. 1985). 

115. W C. V. Bowen. 807 F.2d 1502 (8th Cir. 1987). 



MEDICARE PEER REVIEW 93 

on this program the cumbersome requirements of APA nonce and comment 
rulemaking, with its encrustation of 0MB oversight, would senously impede the 
implementation of this intent, as evidenced by the delay attendant to promulgation of 
the substandard care denial and second opinion rules mandated by COBRA '85. 
which have just reached the NPRM stage.'"" It would be particularly unfortunate if 
both the PRO scopes of work and contracts were forced to go through separate, 
largely redundant rulemaking proceedings that involve four separate Federal Register 
publications. Nevertheless, as has been pointed out, the American Hospital case 
imposes a very expansive reading on the APA exceptions and gives inadequate scope 
to the policies supporting notice and comment rulemaking. 

As was stated at the outset. Congress has had the last word on this question, at 
least for now. Several sections of OBRA '87 attend to earlier recommendations of 
ACUS that HHS take a more public posture in formulating its policies governing the 
Medicare program."' Section 4035(b), for example, provides: 

(2) No rule, requirement, or other statement of policy (other than a national coverage 
determination) that establishes or changes a substantive legal standard governing the scope 
of benefits, the payment for services, or the eligibility of individuals, entities, or 
organizations to furnish or receive services or benefits under this title shall take effect unless 
it is promulgated by the Secretary by regulation . . . ."* 

The amended statute further requires the Secretary to permit at least sixty days for 
public comment on rules so promulgated.'" Under language adopted by OBRA '86, 
this provision does not apply when statutes require implementation of a rule with a 
shorter comment period or within a period of less than 150 days or when the APA 
good cause exception applies. '^o In effect, these provisions explicitly forbid the 
Secretary from revoking HHS's waiver of the APA contracts and benefits exception 
and bind HHS to APA rulemaking when its instructions to the PROs have substantive 
impact. These provisions do not, however, reverse American Hospital, which 
specifically found that the challenged provisions did not have substantive impacts. 
Other provisions of OBRA '87, however, do affect the American Hospital result. 

First, section 4035(c)(1) requires HHS to publish every three months "a list of 
all manual instructions, interpretative rules, statements of policy, and guidelines of 
general applicability" that are promulgated to carry out the Medicare program but are 
not promulgated as rules and have not been published in an earlier list. '2' This 
provision assures that the public will at least be made aware of HHS instructions 
affecting the PRO program. More specifically relevant to PROs, section 4091 



1 16 COBRA "85. Pub. L. No. 99-272. §§ 9401. 9403. 1986 U.S. Com Cong & Admin. News (100 Stat.) 83. 
l%-200. See supra note 61 and mfra Pan VI. 

117 See 1 C F R. §§ 305.86-6. 87-8 (1987). 

118. OBRA '87. Pub. L. No. 100-203. 1987 U.S. Code Cong. & Admin. News (101 Stat.) 1330. 1330-78 (to be 
codified at 42 US C. § 1395hh(a)(2)). 

119 42 use. § 1395hh<bMl)(Supp IV 1986). 

120. Id. § I395hh(b)(2). 

121. OBRA 87. Pub L. No. 100-203. 1987 US. Cooe Cong. & Admin. News (IOI Stat.) 1330. 1330-78 (to be 
codified at 42 U S.C. § 1395hh(cHI)). 



94 TIMOTHY STOLZFUS JOST 

provides that "(tlhe Secretary shall publish in the Federal Register any new policy or 
procedure adopted by the Secretary that affects substantially the performance of 
contract obligations under 42 U.S.C. § l320c-2 (authorizing contracts with PROs) 
not less than 30 days before the date on which such jwlicy or procedure is to take 
effect. "'^2 It further requires that "(tlhe Secretary shall publish in the Federal 
Register the general cntena and standards used for evaluating the efficient and 
effective performance of (PROl contract obligations . . . and shall provide 
opportunity for public comment with respect to such criteria and standards."'-^ 

These provisions make considerable progress toward solving the problem 
addressed by American Hospital. First, they support the court's conclusion that many 
HHS policies and procedures affecting PROs, and specifically the criteria and 
standards for evaluating PRO performance, are not subject to the full panoply of 
procedures that accompany APA rulemaking. The statute does not designate these 
criteria and standards as "rules" subject to the prepublication requirements of 
Executive Order No. 12,291 or No. 12,498 or 5 U.S.C. sections 601 through 612. 
Rather, it simply calls for publication of the criteria and standards in the Federal 
Register and for an opportunity for comment. This should significantly enhance the 
ability of HHS to respond flexibly to the complexity and volatility of the issues raised 
by the PRO program. 

Second, these sections nonetheless affirm Congress' commitment to the pnnci- 
ples that undergird notice and comment rulemaking. HHS should interpret these 
provisions liberally to honor this commitment. In particular, HHS should construe the 
amendments to require publication for comments of proposed scopes of work. The 
scope of work is the most basic document used to evaluate contract proposals and is 
thus ultimately used to judge contract performance. Even though scopes of work are 
arguably, as the D.C. Circuit held, statements of policy rather than rules, they have 
a substantial effect on the PRO program and should be promulgated pursuant to notice 
and comment procedures. 

Drafts of proposed scopes of work have in the past apparently been leaked to 
interested parties before publication. Indeed, a notice of the availability of the Second 
Scope of Work was published in the Federal Register, and HHS received comments 
on it.'-^ HHS would suffer no significant additional burden in sharing them with the 
public generally for comments. With the PRO contract cycle expanded to three years, 
HHS should find that it has ample time to publish proposed scopes of work for 
comment. 

In accordance with the requirements of OBRA '87, HCFA published a summary 
of the Third Scope of Work as a notice in the Federal Register on September 12, 
1988. '-5 This summary was only published, however, after contracts were already 
being negotiated to implement the Third Scope of Work. Moreover, HCFA did not 



122. Id. al 1330-134 (to be codiHed at 42 U.S.C. § l320c-2). This paragraph, like the provisions relating to 
regulations mentioned above, docs not apply when inconsistent with a statutory deadline. 

123. Id. 

124. 50 Fed. Reg. 46.702 (Nov 12. 1985) (comments due by Nov. 27. 1985) 

125. 53 Fed. Reg. 35.234 (Sepi. 12. 1988). 



MEDICARE PEER REVIEW 95 



solicit comments on it. While publication of the summary is a positive step, HCFA 
should do more to involve interested parties in the process of devising scopes of 
work. 

Third, these statutes clarify the status of PROs in their relationship to HHS. The 
statutes reject Judge Wald's p»osition that PROs are mere tools of the federal 
government, no more entitled to an opportunity to participate in policymaking than 
a low level government employee. They recognize rather the quasi-independent 
nature of PROs and their right to notice and an opportunity to comment on the cntena 
under which their performance will be evaluated and to some notice before changes 
are made in their contract obligations. '^^ 

Fourth, these provisions focus on the appropriate level for public participation in 
policymaking. HHS is required to publish policies and procedures "substantially" 
affecting performance of contract obligations and "general" criteria and standards. 
The public should be made aware of and allowed to participate in the formulation of 
broad policy; the minutiae, however, should be left to HHS. In particular, this 
language should be understood not to require publication of individual contracts. The 
intention of Congress in establishing the PRO program, evidenced in 42 U.S.C. 
section 1320c-2, was to run the program through bid contracts. It is difficult to 
conceive of how competitively bid contracts can be subjected to notice and comment 
rulemaking. Moreover, if a scope of work has been run through the notice and 
comment process, there is little need for individual contracts, which reflect the scope 
of work, to be run through a similar process. Despite arguments made earlier that the 
PRO contracts resemble legislative rules, the result arrived at by a combination of the 
American Hospital decision and the OBRA '87 requirements — that the scopes of 
work be published for notice and comment rulemaking, but contracts need not 
be — seems the most workable result. 

Finally, the provisions requiring notice and comment proceedings are rendered 
inapplicable when they conflict with statutory deadlines. Thus, they withhold from 
HHS any excuse for further foot-dragging in complying with statutory mandates. 

IV. Issues Concerning PRO CRrrERiA, Norms, and Standards 

The previous Part considered the rules under which HHS governs the PRO 
program. Each PRO, however, also has its own "rules." by which it governs the 
providers, physicians, and beneficiaries under its jurisdiction. Among the most 
important of these are the criteria, norms, and standards reviewers use to identify 
utilization or quality problems. 

The concepts of criteria and norms have evolved over time. The initial vision of 
the PSRO program was that PSROs would develop objective norms against which the 
performance of providers and physicians could be reviewed. To quote from the 
legislative history: 



126. Ste GIG EfFEcnvENESs Reiiort. supra note 9. at iii. v (concurring with the need for this change). 



96 TIMOTHY STOLZFUS JOST 

The review process would be made more sophisticated through the use of professionally 
developed regional norms of diagnosis and care as guidelines for review activities, a^ 
opposed to the present usage of arbitrarily determined checkpoints. The present review 
process, without such norms, becomes a long senes of episodic case-by-case analyses on a 
subjective basis which fail to take into account in a systematic fashion the expenencc gamed 
through past reviews or to sufficienily emphasize general findings about ihe pattern of care 
provided.'^' 

Such regional norms, under the ongtnaJ PSRO legislation, were to be informed by 
national norms developed by a National Professional Standards Review Council. The 
goal was to develop objective standards of care accepted nationally. '2* 

Though the dream of objective national standards for evaluating the quality and 
effectiveness of medical care continues to inspire scholars.'^' and remains high on the 
research agenda of HHS,'^ PROs have reconciled themselves to the real world where 
such ideal standards do not generally exist. 

To function in such a world, PROs use two kinds of norms for evaluating care. 
First, they use written screening criteria, applied by nonphysician professional record 
reviewers (usually nurses), to identify potential quality and utilization problems. All 
PROs must use utilization, quality, and discharge screens.'^' PROs must use HHS 
generic quality screens for evaluating quality problems. '^^ These screens look for 
problems such as death during or following surgery, nosocomial infection, trauma 
suffered in the hospital, or discharge of a patient with an inappropriately elevated 
temperature, high blood pressure, or depressed pulse. "' PROs are also to apply their 
own cnteria screens to identify premature discharges, unnecessary admissions, 
inappropriate performance of invasive procedures, and, for PPS hospitals, improper 
assignment of DRGs."* Most PROs, including all but one of the PROs with which 
I spoke, use an adapted version of the ISD-A'^' criteria developed by InteiQual for 
making medical necessity and appropriateness determinations."* This system con- 
siders the intensity of services delivered to patients, the severity of the patient's 
condition, the patient's stability at discharge, and the use of special care units and 
clinical support services to determine the necessity and appropriateness of care.'" 



1:7 Senate Comm. on Fw.. S. Ret No. 1230. 92d Cong.. 2d Sess. 257 (Sept. 26. 1972). 

128 See Chenen. PKOs and Poor Quality Medical Care— They Cant Sanction It Until They Define It!. 2 Med 
Staff C<XNS. 25 (Spnng 1988). 

129 Sf. e.g. . Dubois A Brook. Assessing Clinicai Decisio.i Making: Is i.'it Ideal System Feasible?. 25 Inquuiy 59 
(Spnng 1988): Eddy & Billings. The Quality of Medical Evidence: Implications for Quality of Care. '' Health Aff. 19. 
29-30 (Spnng 1988); Wennberg. Improving the Medical Decision-Making Process. 7 Health Aff. 99 (Spring 1988). 

I.V) GAO. supra note 31. ai 144-56 

131 Thuu) Score of Wokk. supra note 3. $ Xill. at 27-29. 

132. Id.. Anach. 1. 

133 Id (Genenc Quaiily Screens. Hospital Inpatient). 

\M Id. § IV. at 3-5 

135 Intensity, seventy, diagnosis — appropnateness. 

136 See AMA. Peeji Review OiicANiZAnoN (PRO Executive) Suhvey Table 8 (1987) (hereinafter AMA PRO 
E.xbx:\.Ti\ i Sl'«vev ); GAO. supra note 3 1 . at 46-5 1 . For an example of PRO screening cnlena. see Black. Impact of PROs 
on Hospiials and Pharmacy Practice. 44 Am. J. Hosr. Pharmacy 77. 79 (Jan. 1987). 

137 Proskctive Payment AssESSMEhtT Comm'n. A Study of the Preaomission Review Process 1-2 (Oec. 1987) 



MEDICARE PEER REVIEW 97 

Second, once problems are identified by nurse reviewers, physician reviewers 
apply their own judgment, using implicit unarticulated standards, to evaluate the 
quality or necessity of the care rendered. ''* It is ultimately on the basis of these 
judgments that PROs deny payment to or sanction providers. Because of the potential 
subjectivity of this system, PROs rely heavily on repetitive review to assure accuracy 
and fairness. An AMA survey, for example, determined that, def>ending on the PRO, 
between three and thirty-five physicians will review a case before an initial sanction 
letter is sent. The mean number of reviewers was ten; the median was seven. '^' Thus, 
although PROs do not apply written cnteria in making iheir final decisions, they do 
seek a consensus medical judgment that transcends the subjective judgment of 
individual reviewers. 

The concepts of criteria, norms, and standards are defined in the PRO statute and 
regulations. 42 U.S.C. section 1320c-3(a)(6) mandates that PROs: 

shall . . . apply professionally developed norms of care, diagnosis, and treatment based upon 
typical patterns of practice within the geographic area served by the organization as principal 
points of evaluation and review, taking into consideration national norms where appropriate. 
Such norms with respect to treatment for particular illnesses or health conditions shall 
include — (i) the types and extent of health care services, which, taking into account 
differing, but acceptable, modes of treatment and methods of organizing and delivering care, 
are considered within the range of appropriate diagnosis and treatment of such illness or 
health condition, consistent with professionally recognized and accepted patterns of care; 
and (ii) the type of health care facility which is considered, consistent with such standards, 
to be the type in which health care services which are medically appropriate for such illness 
or condition can most economically be provided. '*<* 

While the PRO statute speaks only of norms, implicit in it is the distinction made 
by the current PRO regulations among criteria, norms, and standards. Norms are 
defined as "numerical or statistical measures of average observed performance in the 
delivery of health care services."''*' Criteria are "predetermined elements of health 
care, developed by health professionals relying on professional expertise, prior 
experience, and the professional literature, with which aspects of quality, medical 
necessity, and appropriateness of a health care service may be compared. "''•^ 
Standards are defined as "(pjrofessionally developed expressions of the range of 
acceptable variation from a norm or criterion."'"*' In short, norms represent the real, 
criteria the ideal, and standards the acceptable deviations from either the real or the 
ideal.'** 

This neat distinction among norms, standards, and criteria becomes blurred in 
section 466.100, which delineates the use of norms, criteria, and standards. First, 



138. The lenn "standards" is used in a generic rather than a technical sense, as deflned below {see infra text 
accompanying notes 141-44). 

139. See AMA PRO Executive Survey, supra note 136, Table II. 

140. 42 U.S.C. § l320c-3(aM6) (1982). 

141. 42C.FR §466.1 (1987) 

142. Id. 

143. Id. 

144. See A. (JosnEU). PSRO's: The Lav/ and the Health Consumer 34-35 (1975) (further exploring this 
distinction) 



98 TIMOTHY STOLZFUS JOST 

section 466.100(c) requires PROs to "(elstablish wntten criteiia based upon typical 
patterns of practice in the PRO area, or use national cntena where appropnate."'-'^ 
Typical practice patterns, of course, should be reflected in norms, not cnteria. given 
the definitions of the regulations. Section 466.100(a) requires use of "national, or 
where appropnate, regional norms in conducting review to achieve PRO contract 
objectives."'"^ Section 466. 100(b) requires the use of cntena to review patient care 
in health facilities to determine the necessity of admission, of continuing stay for day 
outliers.'*^ or of surgical and other invasive diagnostic and therapeutic procedures, 
and to determine the appropnateness of providing care in particular types of facilities. 
Finally, section 466.100(d) permits the use of vanant cnteria and standards to 
evaluate care in particular locations and facilities if the patterns of practice in them 
are substantially different from practice in the rest of the PRO area and there is a 
reasonable basis for the variation. 

Amendments to the PRO regulations proposed on March 16. 1988, will modify 
the rules in three important respects. First, they make it clear that norms are used by 
nonphysician health care professionals, who screen medical records and refer cases 
that fail the screens to physicians for a final determination. Second, they specify that 
criteria are to be used for reviewing for the quality as well as the necessity and 
appropriateness of norms. Finally, they eliminate the notion of standards (that is, 
acceptable deviations from norms and criteria), noting that PROs in fact have not 
adopted standards.'** 

Several important issues have arisen concerning the application of PRO norms 
and criteria. First, a continuing dispute exists as to whether the norms and criteria 
applied by PROs are sufficiently definite to give adequate notice to those who are 
sanctioned or denied payment by the PROs. Second, differences of opinion exist as 
to what extent local or regional practice variations should be accommodated by PRO 
norms. Third, there is controversy as to what extent PRO utilization norms should 
accommodate social, as opposed to medical, needs for care. Finally, debate continues 
as to what processes should be used to assure the participation of relevant parties in 
the formation of PRO norms and to make final norms available to such parties. 

A. Are PRO Norms and Criteria Sufficiently Definite? 

Administrative law has long struggled with the question of whether standards 
must be prospectively articulated with precision before they can be enforced. There 
is much to be said for administrative agencies governing through precise and 
objective written rules. Objective written standards provide guidance to the regulated 
agencies in planning conduct and thus assure greater voluntary compliance with 



145. 42C.fr. § 466.100(c) (1987). 

146 ll further speciricaily requires ihe use of national norms lo determine (he number of procedures selected for 
preadmission review, a requirement thai would be omitted by (he proposed regulations, as i( does not reflect current PRO 
practice See proposed rule § 466. 1(X). S3 Fed. Reg. 8.660. 8.666 (Mar 16. 1988) 

147. Oulliets are patients who require hospitalization for periods of (ime signincantly longer (han those 
contemplated by PPS and for whom hospitals receive extra reimbursement. 

14«. S3 Fed. Reg. 8.654. 8.660. 8.666 (Mar. 16. 1988). 



MEDICARE PEER REVIEW 99 

requirements. They also focus and facilitate judicial and political review and 
generally cabin administrative discretion.'*' ACUS has long supported prospective 
articulation of agency policies.'^ Courts have on occasion struck down agency 
actions wholly unfettered by written standards. ''■ 

Formulating precise and objective rules, however, may place inordinate de- 
mands on agency resources and may result in rules that are too complicated or that 
are over- or under-inclusive. "^ The Supreme Court has on several occasions 
permitted agencies to make policy on a case-by-case basis rather than through 
prospective rulemaking."' And courts have been notably reluctant to stnke down 
standards propounded by agencies as unconstitutionally vague, at least when neither 
first amendment rights nor criminal prosecutions were involved. '5'» This is particu- 
larly true where industry or professional practice assists in understanding an 
imprecise rule.''' 

The federal statute and regulations defining conduct sanctionable under the PRO 
program are remarkably imprecise. 42 U.S.C. section 1320c-5, for example, 
provides that practitioners or providers have an obligation to assure that services "( 1 ) 
will be provided economically and only when, and to the extent, medically necessary, 
and (2) will be of a quality which meets professionally recognized standards of health 
care.""* Practitioners or providers may be sanctioned if they fail "in a substantial 
number of cases substantially to comply" or in one or more instances "grossly and 
flagrantly" violate these obligations.'" The PRO regulations further define "gross 
and flagrant violation" to mean "(a) violation of an obligation (that) has occurred in 
one or more instances which presents an imminent danger to the health, safety or 
well-being of a Medicare beneficiary or places the beneficiary unnecessarily in 
high-risk situations.""* "Substantial violation in a substantial number of cases" is 
defined as "a pattern of care [that] has been provided that is inappropriate, 
unnecessary, or does not meet recognized professional standards of care, or is not 
supported by the necessary documentation of care as required by the PRO.""' These 
federal standards are fleshed out through objective screening criteria and subjective 
reviewer judgments at the PRO level. 



149. Diver, Agency Articulation of Policy. 1983 Admin. Conf. of the UNfTED States 390-91: Berg, Reexamining 
Policy Procedures: The Choice Between Rulemaking and Adjudication, 38 Admin. L. Rev. 149, 163 ( 1986). 

150 See 1 C F R. § 305.71-3 and other ACUS recominendatioiu cited by Diver, supra note 149. at 389-90. 

151. Morion V Ruiz, 415 US. 199(1974); D& W Food Center v. Block. 786 F. 2d 751, 757 {6th Cir. 1986); White 
V. Roughton, 530 F.2d 750 (7th Cir 1976); Holmes v. New York City Housing Auth.. 398 F 2d 262 (2d Cir 1968) 

1S2 Diver, supra note ;40. at 397-98. 505-06. 508-09 

153. NUtBv. BellAerospaceCo , 416U S 2o7 (1974); NLRB v. Wy man Gordon Co , 394 US 759(1969) Sr*- 
Oiso ScaJia, Back to Basics: Making Law Without Making Rules. 5 Regulation 25 (July/Aug. 1981). 

154 United States v Sun Sport Imports Ltd.. 725 F 2d 184 (2d Cir. 1984); Vorbcck v Schmcker. 660 F 2d 1260 
(8th Cir. 1981), cert, denied. 455 U.S. 921 (1982). See Village of Hoffman Estates v. The Flipside. Hoffman Estates. 
Inc.. 455 US 489. 498 (1982). 

155. PBR Inc V Secretary of Labor. 643 F.2d 890 (1st Cir. 1981); Brass Plating Co v Town of Windsor. 639 F 
Supp. 873 (D. Conn. 1986) 

156. 42 use. § 1320c-5(a)(l)-(2) (1982). The section further requires that providers and practitioners be able to 
supply evidence of the necessity and quality of services provided. 

157 Id. § 1320c-5(b)(l)(A). (B). 

158. Id. 

159. 42C.FR. § 1004 I (1987). 



100 TIMOTHY STOLZFUS JOST 

The federal PRO statute and regulations have been upheld repeatedly against 
vagueness challenges. '*** The Fourth Circuit in Varandani v. Bowen noted: "The 
definition of adequate medical care cannot be boiled down to a precise mathematical 
formula; it must be grounded in what, from time to time, other health professionals 
consider to be acceptable standards of health care. "">' The First Circuit Court of 
Appeals in Doyle v. Bowen further noted that "|t|o the medical profession, which 
will administer this (federal) standard, it has a reasonably clear meaning."'" 

Nonetheless, physicians sanctioned under the Medicare program, and many of 
the provider attorneys with whom I spoke, have complained about the lack of 
precision in PRO criteria for evaluating the necessity and quality of medical care. '" 
While ALJs hearing PRO sanction cases have not rejected PRO sanctions solely 
because of the imprecision of PRO standards, they have called for more objective, 
clear, and appropriate standards.'^ On the other hand, some provider associations 
that I interviewed expressed continuing concern about reducing the practice of 
medicine to a "cookbook," which ignores the art necessarily involved in medical 
practice.'" Devising a definitive and precise formula for evaluating medical practice 
may be an impossible dream in any event — a recent article that proposed use of 
Bayesian analysis for devising criteria maps for analyzing patient diagnosis noted that 
a complete system would include over ten billion pathways of analysis for common 
medical problems.'^ Some room must be left for informed judgment. 

HHS and the PROs continue to attempt to devise more objective critena for 
evaluating the effectiveness and quality of medical care. The generic screening 
criteria uniformly used by the PROs represent one such attempt. Reviews of these 
criteria are generally positive."*' HHS has also proposed that more explicit cnteria be 
developed for particular conditions before the denial of payment for substandard care 
program, mandated by COBRA '85, is implemented.'** At the PRO level, the 



160. Doyle v. Bowen. 84« F.2d 2%. 301 (1st Or. 1988): Varandani v Bowen. 824 F 2d 307. 31 1-12 (4ih Cir. 
1987); Associanon of Am. Physicians & Surgeons v. Weinberger. 395 F. Supp. 125. 138 {N.D. 111. 1975). 

161. Voraikiafu. 824 F.2dai 312. 

162. Doy/«.848F.2dai30l:5««a/5o/nr«Lirshutz. No. 000-44-7020. 1 1-13 (Mar. 25. 1988) (AU opinion holding 
that PROs need not apply wrinen cnicfia. norms, and standards in sanctioning physician). 

163. The interviews conducted by the author were generally earned out under an understanding (hat (he identiiy of 
interviewees would be kept confidentiaJ. so interviewees are not idennried here. For published sutemenu to the same 
effect, see generally Cheiten, supra note 128; Hearings on the Peer Review Organization Program Before the Subcomm 
on Health and the Env't of the House Energy A Commerce Comm. , 100th Cong.. 1st Sess. 7-9 ( 1987) (statement of 
AHA). 

164. /»ir*Gieene. No. HlX-00^00-0219, 8-20 (Dec. 29. 1987);//. r* ApakUj)akul. Nc. HIX-000-34-7009. 32-33 
(June 1. 1987) (charges too vague). 

165. See Interpreting PSRO "NormsofCare" — Cookbook Medicine or Textbook Medicine? . Am. Med. Ntws. May 
20. 1974, at 21; PROs. Medical Societies Musi Play a Role. The Itn^RNisi. Apr. 1985. at 9 

166. Dubois & Brook, supra note 129. at 63. 

167. See OmoE of Inspecto* Genekal, U.S. Dep't of Health & Human Seavs., The Utiuzation and Quality 
CotOTtOL Pe£« Review Organization (PRO) Program. QuALmr Review AcrivmES 12 (1988) (hereinafter DIG QuALmr 
REKNtT); Slide Presenution on (Quality Assurance. Ernst & Whinney. Slide K-18 (1987 survey results). 

168. Health Car£ Fw. Aomin.. U.S. Dtp'rof Health & Human Servs.. Reculahons on Denial of Payment for 
Substandard QuALTTY Care (draft) A12-13 (Feb. 29. 1988) (hereinafter HCFA Regulations!; Proposed Substandard 
Ca*e Regulations, supra note 61. at 1958-59. 1963 (to be codiHed at 42 C.F R. pt. 462. § 466. 100(c) (establishment 
of cnteria and standards)). 



MEDICARE PEER REVIEW 101 

Pennsylvania PRO has been expenmenting with computenzing its screening cntena, 
which presumably would require greater precision. 

HHS and the PROs should continue to press forward toward formulating more 
objective and precise necessity and quality cntena when it is possible to do so with 
the support of scientific evidence and the consensus of the medical community. Until 
such cntena become feasible, however, the program will continue to have to live with 
relatively imprecise cntena that are applied to individual cases through the judgment 
of physician reviewers. 

B . Local or National Norms 

One of the most politically controversial issues affecting the PROs has been 
whether norms and criteria used for evaluating medical care should be national in 
scope or whether local practice variations should be accommodated. From 1985 to 
1987, 63% of physicians recommended by PROs for sanctions and 65% of the 
physicians sanctioned by the OIG were from rural areas, despite the fact that only 
1 1% of physicians in the United States practice in rural areas.'*' While there is no 
consensus as to why this has been true, it has clearly raised the ire of rural 
practitioners and their patients, who believe that rural physicians are sanctioned 
disproportionately because PROs depend on reviewers from urban areas who do not 
understand the practice of medicine in rural areas.'™ 

It has long been known that there are substantial variations in medical practice 
between different communities,'^' a fact recognized, at least intuitively, in the old 
malpractice locality rule. On the other hand, there is a certain plausibility to the 
argument (which has largely won the day in malpractice litigation) that local 
standards often reflect poor quality care and that a national standard, adjusted for 
local variations in access to medical resources, is more appropriate. A broken hip or 
heart attack in rural Texas is, after all, physiologically identical to one in Boston. "^ 

From the beginning, the PRO statute recognized that there may be local or 
regional variations in acceptable norms of treatment and that the PROs should take 
these into account. The PRO regulations permit the use of variant norms to 
accommodate local variations in practice."' PRO Interim Manual Transmittal 
IM-87-1'''* requires PROs, whenever possible, to use physician reviewers who 
practice in settings similar to that of the physician under review and specifically to use 



169. OIG Sanction Report, supra nwe 6. App. XI. Table D. See id. a( 20 (speculating as (o the reasons for this). 

170. See Hearings on Peer Review Organizations. Health and the Env'l Subcomm. of the House Energy and 
Commerce Comm.. 100th Cong., I si Sess. 30 (1987) (statement of Rep. Boulter); Reinhold. Quality-Care Reviews 
Hurling Rural Doctors. NY. Times. Mar. 24, 1987, at Al. col. 3. 

171. Wennberg. Dealing with Medical Practice Variations: A Proposal for Action. 3 Health Aft 6 (Summer 
1984). 

172. See Hall v Hilbun. 466 So. 2d 856 (Miss. 1985); letter from John C. Rother. for ihe Am Assn of Retired 
Persons, lo William Roper, Administrator. HCFA, at 5 (May 16. 1988) (commenting on March 16, 1988. proposed PRO 
regulations) ("AARP firmly believes that the standard of care must not change from location to location, although 
allowance may be made for resource availability."). 

173. 42 C F R. § 466.100(d) (1987). 

174. PRO Manual, supra note 53, § 3015 (Apr. 1987). 



102 TIMOTHY STOLZFUS JOST 

rural physicians to review the work of rural physicians."' At least one court has 
determined that a physician who argued that a PRO must use rural standards m 
sanctioning a rural physician raised a sufficiently substantial question to secure a 
preliminary injunction.'^* This interpretation of the PRO legislation was followed by 
the ALI in that case on remand.'^' Finally, OBRA "87 further accommodates local 
practice variations by requinng PRO consideration of the "special problems 
associated with delivenng care in remote rural areas'*'^* and by mandating on-site 
review of at least 20% of the rural hospitals in a PRO's area.''' 

In practice, PROs are increasingly sensitive to the concerns of rural practition- 
ers. None of the PROs I interviewed had special criteria sets for regional areas, but 
all tried to use rural reviewers for rural providers and physicians. The OIG sanction 
activity investigation found that 63% of those interviewed believed there was no bias 
against rural physicians and providers in PRO quality review activities.'*** 

While substantial political and, perhaps, medical arguments can be made both 
for and against the use of local or national standards, little can be said about the issue 
from an administrative law perspective. One point, however, should be made (and 
here this Article briefly gets ahead of itselO- OBRA '87 provides that before HHS 
can exclude a provider or practitioner located in a rural health manpower shortage 
area or in a county with a population of less than 70,000 from the Medicare program, 
the provider or practitioner must be afforded a hearing, at which HHS must prove by 
a preponderance of the evidence that the provider or practitioner will pose a serious 
risk to beneficiaries if allowed to continue in the program. '*' As of late October 1988 
three exclusions appealed under this procedure had been stayed (two by consent of the 
OIG), and two others remained pending.'*^ None of the contested cases has been 
decided within forty-five days. Thus, prehearing exclusions of rural practitioners 
from the program have effectively halted. 

The appropriate procedure for sanctioning physicians or providers is an 
important issue and will be discussed below. Whatever procedure is appropriate, 
however, fundamental constitutional notions of equal protection and administrative 
law principles of fairness demand that the same procedure be applied both to rural and 
urban practitioners. There are arguments, of course, for applying more favorable 
procedures in rural areas: there is the charge of PRO bias against rural practitioners 
and the concern that beneficiaries in rural areas may have no alternatives if the sole 



175 Proposed rule § 406.98 also adopts 'his policy 53 Fed. Reg 1».665 (Mar. 16. 1988). A receni GAG report 
found thai only about half of PRO reviews of cases handled by rural physicians were in fact reviewed by rural practitioners 
at the ininal review stage. GAG, Meoktaiie PROs: Extkeme Vajuations in GitCAMiZAnoNAL Structure v<d AcnvmES 21 
( 1988) (hereinafter GAG Meoktare PRGs| 

176 Greene v Bowen. 639 F Supp. 554 (E.D. Cal. 1986). 

177 In rf Greene. No. HIX-000-00-0219. at 15. 20 (Oec. 29. 1987). 

178 OBRA '87. Pub. L. No. 100-203. § 4094(a). 1987 U.S. Cooe Cong. & Admw. News (101 Sut.) 1330. 
1330-136 

179 Id § 4094<b). 1987 US. Cooe Cong. & Admin. News (101 Sui.) 1330. 1330-137 
180. OIG Sanction Rejiort. supra note 6. at 19. 

181 42 use. § l320c-5(bK5)(1982). 

182. Telephone interview with William Ljbercci. Offlcc of Inspector General. U.S. Dep'i of Health & Human 
Servs. (Get. 31. 1988). 



MEDICARE PEER REVIEW 103 

provider or practitioner in the area is excluded from the Medicare program. But the 
charge of bias has been generally rejected, and potential bias problems can be 
addressed through less drastic means, such as requinng the use of rural physicians to 
review care delivered in rural areas. The OIG already considers "the availability of 
alternative sources of services in the community"'** in determining whether or not to 
exclude a provider or practitioner. Neither problem is senous enough to justify 
supplying rural practitioners and providers with radically different procedural 
protections than those offered others.'** Congress, therefore, should repeal this 
provision. 

C. Consideration of Social Criteria 

One of the major tasks of the PRO program is "appropriateness" review — 
determining whether care provided on an inpatient basis could have been provided 
more appropriately through less expensive outpatient treatment.'" A procedure 
generally appropriate for outpatient care considering only medical factors may in a 
particular case be inappropriate because of the social characteristics of the patient or 
situation. One interviewee, for example, recounted the experience of a patient who 
traveled from her home in rural North Dakota several hundred miles to Minneapolis 
to be hospitalized for a medical procedure, only to be told on arrival that she would 
have to leave the hospital immediately after the procedure was done because the PRO 
would only permit it to be done on an outpatient basis.'** Section 4094(a) of OBRA 
'87 provides that PROs: 

shall take into account the special problems associated with delivering care in remote rural 
areas, the availability of service alternatives to inpatient hospitalization and other appropn- 
ate factors (such as the distance from a patient's residence to the site of care, family support, 
availability of proximate alternative sites of care, and the patient's ability to carry out 
necessary or prescribed self-care regimens) that could adversely affect the safety or 
effectiveness of U^atment provided on an outpatient basis. '"^ 

While there are substantial arguments for the consideration of social factors in 
determining the appropriateness of outpatient care (or, for that matter, in determining 
the appropriateness of hospital discharges), implementation of this provision will 



183. 42CFR. § 1004 80(0(1987). 

184. The HaJI Amendmenrs ongin in (he frustnuion with and anger toward the Inspector General felt by some 
physKians a welt illustrated by remarks of iu diafter. Rep. HaJI. responding to the first use of the procedure to st;iy on 
exclusKMi: 

This is one of the Tirst judicial victoi les and should indicate to the inspector geiwral. kiciiard Ku:kSeiow. that 
he can't sit over there in his ivory castle and make all types of blustery and bragging and misleading speeches; 
thai while he can talk down to men and women who come before him and have heretofore been at his 
mercy — that there are many Members of Congress who are neither interested in nor impressed by his pomposity . 
Even physicians have a right to due process — and in the short 120 days you have left — Mr. Kusserow — I 
suggest you try to get acquainted with this fact, and never again try to deny any American due process! 
134 Cong. Rec. E2803-04 (daily ed. Aug. 1 1 . 1988) (extension of rentarks by Rep. HaU). 

185. 42U.S.C, $ l320c-3(aKIMC)(l982). 

186. See also Hearings on Medicare Peer Review Orgs.. Subcomm. on Health and the Env'l. House Energy and 
Commerce Comm.. 100th Cong.. Ist Scss. 12-15 (1987) (sutement of Viu Ostrander. AARP) |hetcinafter AARP 
Suiement). 

187. OBRA "87. Pub. L. No. 100-203. 1987 US. Cooe Cong. &. Admin. News (101 Stai.) 1330. 1330-136. 



104 TIMOTHY STOLZFUS JOST 

obviously result in more inpatient care and. thus, higher expenditures for the 
Medicare program. Whether such added expenditures are justified is a policy 
question, not a question of admmistrative law. HCFA implementation of this 
provision, however, does raise an important administrative law issue. 

In implementing OBRA '87, HCFA initially took the position that the first 
comma in the section quoted above should be read as a colon, that is. that the 
provision only applied to remote rural areas. This is not only contrary to the 
amendment's clear language, but is also contradicted by the legislative history of the 
provision, which clearly states that beneficiary location in a remote rural area is only 
one of the social factors to be taken into account.'** HHS is apparently reconsidering 
its position on its interpretation of section 4094(a) and should expeditiously 
implement it as written.'** 

D. Formulation and Publication of Criteria and Standards 

As has already been briefly noted, the D.C. Circuit held in Public Citizen Health 
Research Group v. DHEW that PSROs were not agencies within the meaning of the 
FOLA."" Because the FOIA adopts the APA definition of "agency," that decision 
also stands for the proposition that PROs are not agencies under the APA and thus not 
subject to the rulemaking requirements of 5 U.S.C. section 553. Although the 
reasoning of the court in Public Citizen is somewhat confused (the court held both 
that the PSRO did not have independent decisionmaking authority and that it was not 
subject to supervision by HHS).'" its holding seems consistent with congressional 
intent that PROs are to be independent bodies of practicing physicians and not 
government agencies. "^ The holding is also consistent with other cases that have 
considered the status of PSROs and PROs. "^ PROs are also probably not subject to 
the Model State APA, which describes an agency as "a board, commission, 
department, officer, or other administrative unit of this state.""* 

Even though PROs are themselves not bound by the rulemaking requirements of 
the APA, the reasoning behind those requirements'" argues in favor of PROs seeking 
participation of affected parties in formulating criteria, standards, and norms and 
assuring that their criteria, standards, and norms are made available to those affected 
by them. This is recognized by the documents governing the PRO program. The 
Third Scope of Work requires that: 



188. H.R. CoNF. Rtp. TO Accompany rwE Omnibus Budget Reconouation Act of 1987. lOOih Cong.. Isi Scss. 
661. 663 

1 89. Several beneficiary rcpresentalives with whom I spoke also argued for grcaier consideration of social cniena 
in determining appropnateness of discharge. This is. however, a policy and not an administrative law issue, and it is thus 
beyond the scope of this Article. 

190. 668 F.2d 537 (D.C. Cir 1981). 

191. Id. at 543-44 

192. See id. at 542-43. 

193. See Oster v Bowen. 682 F. Supp. 853 (E.D. Va. 1988); Kwoun v. Southeast Mo PSRO. 663 F Supp. .s:0 
(DC. Mo. 1985); Taylor v. Rint Osteopathic Hosp.. 561 F Supp. 1152 (E.D. Mich. 1983). 

194. Model State Admin. Procedure Act § 1-102(1) (1981). 

195. See supra Part III for discussion. 



MEDICARE PEER REVIEW 105 

In the development/refinement of cntena, ihe PRO shall consull with physicians/practiiion- 
ers actively engaged in practice in the state. PRO developed cntena are to be submitted to 
HCFA for review, and sent to the providers and physician organizations (i.e.. State medical 
societies) in the State. In addition, the contractor must furnish the criteria to anyone upon 
request.'*** 

The PRO regulations also require that a PRO must disclose "(tjhe norms, cinteria, 
and standards it uses for initial screening of cases, and for other review activities. "'^^ 

All of the PROs with which I spoke attempt to involve their state medical 
community in drafting or modifying their criteria sets. Most PROs seek comments 
from state medical associations and specialty societies; some circulate drafts to 
hospital medical staffs or have provider advisory committees. Seventy-three percent 
of state medical societies refxjrted in an AMA survey that state specialty societies had 
assisted PROs in developing or modifying criteria, and 39% reported that specialty 
sections of the state medical society had also provided assistance.'^* The possibility 
of PROs involving their own state medical community in drafting and reviewing PRO 
criteria is one of the strongest arguments against the creation of uniform national 
criteria sets by HCFA. 

All of the PROs that I interviewed are also making efforts to publicize their 
criteria. All send their criteria sets to every hospital in the state. Some have identified 
PRO liaisons in the hospitals to whom they direct the criteria, others send them to the 
director of the medical staff, the administrator, or the utilization review/quality 
assurance supervisor. Most also send them to the state medical society. None sends 
its criteria to all physicians in the state, though some send out a regular bulletin or 
newsletter to advise doctors of criteria changes or send specialty criteria to affected 
specialists. Nevertheless, several state provider representatives with whom I spoke 
said that providers are not sufficiently educated by the PROs as to PRO criteria. "^ 
National provider and beneficiary groups have also called for clearer and more public 
articulation of criteria. ^oo 

The task of formulating PRO criteria is ideally suited for a simplified version of 
negotiated rulemaking. 2*" The parties affected by criteria are relatively few, and their 
spokespersons are readily identifiable; technical rather than ideological issues are 
usually (though not always) involved; the adoption of some criteria is usually 
necessary and inevitable; and opportunities for compromise and consensus are usually 



196. Third Score of Won-., supra noce 3. § XIll, at 27-78. 

197. 42C.F.R. |476.l20(aHI)(l987). PROs mu:it also make available to the public copies of their contracts (which 
descnlx their procedures) and copies of documents describing their administrative procedures. Id. § 476. 120(a)(2)-<3). 

198. AMA PRO ExECvmvE Survey, supra note 136. at 2. 

199. A recent study by ihe GAG provides more information on how PROs disseminate their criteria. See generally 
GAG Medicare PRGs, supra note 175. 

200. See Hearings on Medicare PROs Before Subcomm. on Health and the Env't. House Comm. of Energy and 
Commerce. 100th Cong.. 1st Sess. 7-11 (1987) (statement of AHA); Information for Consumers About Quality of 
MedicalCare. Subcomm. on Natural Resources, Research. andEnv'i, HouseComm. onScience. Space and Tech. . 100th 
Cong.. 2nd Sess. 10 ( 1988) (statement of AARP); AARP Statement, supra note 186. at 4. 

201. See Harter. Negotiating Regulations: A Cure for Malaise. 71 Geo. L.J. I (1982); Perritt. Negotiated 
Rulemaking Before Federal Agencies: Evaluation of Recommendations by the Administrative Conference of the United 
States. 74 Geo. L.J. 1625 (1986); I C.F.R. §§ 305.82-4, .85-5 (1987). 



106 TIMOTHY STOLZFUS JOST 

present. '02 Greater involvement of providers and practitioners in the standard-setting 
process could bring greater understanding of the standards and compliance with them, 
acceptance of the legitimacy of PRO review, and. most importantly, better standards. 
Greater involvement of beneficiary organizations would assure that the patient's 
perspective on care was taken into account, especially when social issues are 
involved. Dissemination of norms to beneficiaries could also play a valuable 
educative role.^^' PROs establishing or modifying critena should, therefore, convene 
groups of provider and beneficiary organizations and attempt to reach agreement as 
to appropriate criteria through a negotiating process. Such an approach to criteria 
setting is in the best tradition of peer review. 

Two provisions of OBRA '87 have the potential to make an important 
contribution to the task of educating the medical community about PRO criteria and 
norms. Section 4094(c)(l)(B)(i) requires PROs to offer to make available a physician 
to meet with the medical and administrative staff of each hospital to explain the 
organization's review of the hospital's services (at individual hospitals or on a 
regional basis). ^o* Section 4094(c)(l)(B)(ii) requires PROs to publish at least 
annually and distribute to practitioners and providers a report describing the types of 
cases in which the PRO has frequently determined that care has been provided 
inappropriately, unnecessarily, or not in conformity with professional standards of 
care. 2°' These provisions are implemented by the Third Scope of Work, which also 
requires PROs to engage in further efforts at provider education, including making 
staff available for educational presentations, and publications describing PRO 
procedures. ^'^ Educational efforts that involve direct discussions with affected 
providers and practitioners and focus on particular problem areas have great promise 
for effectively conununicating PRO criteria. 

V. The pro Quauty Assurance and Sanction Program 

A. Introduction 

By far the most controversial function of the PRO program has been its 
sanctioning authority. 42 U.S.C. section l320-5(a) imposes on practitioners and 
providers participating in the Medicare program an obligation to assure that services 
they render are provided economically, are provided only when medically necessary, 
and are of a quality that meets professional standards of care. It further obligates them 
to provide evidence to establish that they have met these requirements. It also gives 
PROs the power and responsibility to sanction providers who fa.l to fulfill these 
obligations. If "after reasonable notice and opportunity for discussion" a PRO 
determines that practitioners or providers have "(A) failed in a substantial number of 
cases substantially to comply" with these obligations or "(B) grossly and flagrantly 

202. See Perrin. supra note 201. at 1642-47; I C.F.R. J 305.85-5 (1987) 

203. See A. Gosreld, PSRO's: The Law and the Health Consumer 32-33 (1975). 

204 OBRA 87. Pub. L. No. 100-203. 1987 U.S. Cooe Cong. A Admin. News (101 Star) 1330, 1330-136 

205. Id. at 1330-137. 

206. Third Scope of Work, supra no«e 3. § XVIII(C). at 40-41. 



MEDICARE PEER REVIEW 107 



violated any such obligation in one or more instances," the PRO shall submit a report 
and sanction recommendation to the HHS OIG.-°^ If the OIG agrees with the PRO 
recommendation and determines that the practitioner or provider "has demonstrated 
an unwillingness or a lack of ability substantially to comply with such obligations." 
it may exclude the provider or practitioner from "eligibility to provide Medicare 
services on a reimbursable basis. "^°* When the provider or practitioner has been 
sanctioned for providing medically improper or unnecessary care, the OIG may, 
instead of exclusion, impose a monetary penalty not in excess of the cost of the 
medically improper or unnecessary services. '^^ A sanctioned provider or practitioner 
may obtain a p>ost-exclusion hearing before an ALJ and, ultimately, judicial review. 
As was discussed earlier, under OBRA '87 rural providers cannot be excluded from 
the program without a pre-exclusion hearing before an ALJ to determine whether the 
provider or practitioner poses a serious risk to Medicare beneficiaries. ^'<* 

B. Criticisms of the PRO Sanction Process 

The PROs' exercise of their sanctioning authority has made them the target of 
much criticism. As of December 3 1 , 1987, 38 of the 54 PROs had referred 151 cases 
to the OIG; the OIG had excluded 60 physicians and 1 facility and assessed a penalty 
against 24 physicians and 2 facilities. 2" This represents about 19 sanctions per 
100,000 physicians serving Medicare beneficiaries. Through September 1987 
twenty-three states and the District of Columbia, containing one-fifth of the nation's 
doctors and Medicare beneficiaries, had not successfully recommended a single 
sanction. ^'2 Though it is impossible, of course, to determine the optimal level of 
sanctioning, it is difficult to believe that 99.98% of doctors treating Medicare 
beneficiaries (and 100% in twenty-three states) are doing a satisfactory job, especially 
given common estimates that incompetency among doctors runs at levels of 5% to 
10%.2'3 Some consumer groups have been sharply critical of the timidity of the PROs 
in sanctioning doctors, ^'^ and the OIG has occasionally joined this criticism. ^'5 

Attorneys representing providers with whom I spoke, on the other hand, are 
sharply critical of what they see as oppressive aggressiveness on the part of the 



207. 42 use. 8 I320c-5(b)(l982). 

208 td. § I320c-S(b)(l)- This provision funher sutes thai if HHS fails to take action within 120 days, the 
ptaciiiioner or provider wilt be excluded until HHS decides otherwise. This clause was put in the statute to addrc's the 
proMem u( HHS delay in acting on sanctions recommended by t^ie PSROs. The OIG has consisien'.ly taken action on PRO 
recommendation:; witnin 120 days, consigning this provision to desuetude. 

209. Id. S I320c-5(b)(3). 

210. Id. § l320c-S(bKS). HHS has implemented these statutory provisions through regulations appearing at 42 
C.F R. pi. 1004 and PRO Mamual. supra note 53, Transminal No. 15 of May 1987 

211. See GIG Sanction Rei>oht, supra note 6, at II. 

212. Subcomm. on Health. Comm. on Ways and Means. lOOth Cong., tst Sess. 2, 3 (1987) (testimony of Sidney 
Wolfe. Director Pub. Citizen Health Research Group) (hereinafter Testimony of S Wolfe). 

213. See [Derbyshire. Medical Discipline in Disarray: The Incompetent Physician. 18 Hosp. Prac. 30, 31 (Nov. 
1983). 

214. Testimony of S. Wolfe, supra note 212; Kotelchuck. Watchdog on a Short Chain. Healto/PAC Bun... Spring 
1987, at 19. 21. 

215. See Webber. A PRO Report Card. The Intbknist. July 1986. at 7. 9. 



108 TIMOTHY STOLZFUS JOST 

PROs.-'o They properly note the devastating effect that exclusion from the Medicare 
program has on physicians. While an excluded physician may in theory continue to 
practice (and even to treat Medicare beneficiaries if he or she is willing to do so 
without compensation). Medicare nationally pays for 21% of physicians services^ '^ 
and provides a much higher proportion of the income for some specialists. Secondary 
effects of Medicare exclusion can, moreover, be even more devastating. Attorneys 
with whom 1 talked stated that after exclusion their clients ceased to receive referrals 
from other doctors, were terminated from hospital medical staffs, and were 
investigated by slate licensure agencies. While they waited up to eighteen months for 
ALJ hearings and judicial review, excluded doctors were forced to maintain the high 
overhead of their practices and pay high legal fees without income, an ordeal that 
drove at least one into bankruptcy. 

The proceedings that resulted in these consequences were passionately criticized 
by provider representatives. Though PRO procedures were substantially amended 
under a May 1987 compromise between the OIG, HCFA, the AMA, and the 
American Association of Retired Persons (AARP) to assure greater procedural 
protections for sanctioned providers and practitioners at the PRO level, ^'* provider 
attorneys still complain of the inability to cross-examine the reviewers (often 
anonymous) who initiated the proceedings, the informality of the proceedings, and 
the lack of notice and standards. Their harshest criticism, however, is focused on the 
perceived bias of PRO sanction proceedings. Physician representatives widely 
believe that PROs are directly rewarded for sanctioning providers and threatened with 
contract termination for not doing so, and thus are heavily biased against providers 
once sanctions are recommended. They also believe thai the OIG is prejudiced 
against providers and does not provide an effective review of PRO recommen- 
dations. ^'^ 

The vast majority of those involved in the sanction process, including the PROs 
and many beneficiary and provider representatives, are somewhere between these 
poles and most show a marked ambivalence about PRO sanctions. On the one hand, 
87% of respondents interviewed in a recent study felt that the PROs' sanction 
authority strengthened their ability to perform their mission. ^^o Without the threat of 
sanctions, the PROs' educational and corrective efforts would be less likely to get the 
attention of the medical community. On the other hand, the PROs feel uncomfortable 
with the hostility the sanction process has generated and are frustrated by the time and 
resources that go into the sanction process. Two-thirds of PRO CEOs interviewed by 
the OIG felt that the sanction process continues to be problematic."' PROs are 



216. See also Schorr. Peer Review: Siill Tilling at Windmills. 27 Phys. Mem. 124 (Oct. 1987); Remaris by 
Hanson. Cal Med Assn 1987 House of Delegates (Mar 10. 1987) 

217 Div of Nail Cost Eslimaies, Office of ihe Actuary. Health Care Fin. Admin.. National Health Expenditures. 
I9H6-20OO. 8 Health Care Fin Rev . Summer 1987. at 1. 13 

218 See US. Dep I of Health & Human Servs News Release. Changes in PRO Sanction Process. May 13. 1987 
(implemented by Manual Transmittal 15). 

219 See Carlova. Have Peer Reviewers Put a Price on Your Head?. Med. Econ , Sept. 5. 1988. at 62. 
220. OIG Sanction Report, supra note 6. at 12. 

221 /</. at 15. 



MEDICARE PEER REVIEW 109 

particularly frustrated with the inconsistent instructions they get from HCFA and the 
OIG as to when to sanction, the proportion of PRO sanction recommendations 
rejected by the OIG (over half in 1987)."2 and their rate of reversal in AU 
proceedings. 

One thing is clear the rate of sanctioning activity has dropped dramatically m 
the recent past, after climbing steadily since the inception of the PRO program. 
Dunng 1985 seven cases were recommended to the OIG by the PROs for sanctions 
and all were accepted. During 1986, the high water year for sanction activity, 
seventy-seven were recommended, and fifty-five were accepted. During 1987 the 
level of recommendations dropped slightly to seventy, but the number of OIG 
acceptances dropf)ed dramatically to twenty-five.^^ As of July, only ten cases had 
been referred to the OIG for sanctions in 1988. With individual PROs, the drop-off 
has been even more dramatic. In the first two years of the program, the California 
PRO was among the most active in recommending sanctions, but within the last year 
it has not referred any cases to the OIG for sanctions. 

The reasons for this decline in activity are disputed. The most optimistic view 
is that the initial flurry of sanction activity has cleared up all problems, and further 
sanctions are not necessary. A more realistic view is that the PROs. having gotten the 
attention of physicians, can now rely more on less drastic interventions. Additional 
procedures imposed by the May 1987 compromise may have impeded the sanction 
activity of some PROs, but many of the PROs had already implemented these 
protections prior to the compromise. The most likely explanation for the slowdown 
is that a variety of administrative impediments are discouraging the PROs from 
sanctioning the physicians. First, the OIG has rejected a large proportion of sanction 
referrals, which has discouraged PROs from referring cases. Second, PROs have 
noted the high rate of ALJ reversals of sanction cases and have become reluctant to 
conrunit the heavy expenditure of resources necessary to get a case to the ALJ level 
if reversal seems inevitable. 

In order to understand the decline in PRO sanction activity, it is important to 
consider the reasons why PRO sanction cases are being rejected. Of the fifty-six cases 
rejected by the OIG in 1987, twenty were rejected because the PRO had failed to 
establish the inability or unwillingness of the physician or provider to meet statutory 
obligations. "4 PROs have generally concluded that they will not be able to establish 
inability or unwillingness without documenting a pattern of violations over a penod 
of time, and they are thus keeping physicians under correction plans for several weeks 
or monttis befoie referring for sanctions. Many PROs have also concluded that the 
OIG will not accept referrals when only a single instance of gross and flagrant 
conduct is involved (despite clear statutory language to the contrary) or of cases 



222. Id. al 17. 

223. W. 

224. Id. at 18. 



1 10 TIMOTHY STOLZFUS JOST 

involving fines. ^" PROs have also expenenced difficulty collecting payments from 
HHS for sanction activities, which are supposed to be separately reimbursed.--'' 

In sum. despite significant changes in the sanction process over the last two 
years, the process still seems not to be working very smoothly. 

C. The PRO Quality Assurance and Sanction Process Described 

All PROs. as required by the Scope of Work and PRO contracts, have a 
two-stage process through which cases must go before sanctions are imposed. 
Sanction cases normally begin in the quality assurance process, which identifies and 
attempts to correct quality problems.^" Senous or repeated problems identified 
through the quality assurance process are sent on to the sanction process. The Third 
Scope of Work establishes broad guidelines for the quality assurance process; the PRO 
regulations^^ set out more specific guidelines for the sanction process. Both 
processes vary considerably, however, from PRO to PRO. 

In all PROs. the first step down the long road to a PRO sanction is taken (in most 
instances) when a nurse reviewer kicks out a case for failing one of the generic quality 
screens or another quality, utilization, or discharge screen. ^^ In most PROs, the case 
is then reviewed by a physician reviewer, who will in all likelihood not belong to the 
specialty of the doctcu* under review. If the review is conducted on-site at a hospital, 
the doctor is even less likely to be specialty-matched. ^^ 

If the initial physician advisor determines that there is a quality problem, one of 
several things may happen. In four of the twelve PROs I interviewed, the PRO sends 
a letter to the attending physician or provider and asks for an explanation of the 
problem. In six of these PROs, the case is referred to one or more specialists matched 
with the specialty of the physician under question, who further reviews the case. If 
the specialist agrees that there is a problem, the attending physician or provider is 
contacted for further information; otherwise the matter is dropped. Finally, in the two 
other PROs with which I spoke, the case is sent to a medical director or quality review 
committee simultaneously with or before a request to the attending physician or 
provider for more information. In several of the PROs, minor problems are noted at 
this level for futtire consideration if patterns develop, but they are not immediately 
acted on. 

Once a response is received from the attending physician or provider, the 
response is reviewed, in most of the PROs, by a matched sjjecialist, usually someone 



225. Id. al 16. These issues are the subject of a recetu GAG invesiigalion. See generally GAG Mfdktau PROs. 
supra ncMe 175. 

226. GIG SANcnoH Rekwt. supra note 6. at 16 Paymem for sanamn activity after an initial sanction lener is sent 
IS supposed to be in addition to the amount included in the PRO coninci for quality assurance generally See 2 Medicare 
& Medicaid Guide (CCH) 1 12.875 (1986). 

227. Potential sancoon cases can also be identiried in the utilization review process, but they seldom are. 

228. 42C.fr p« 1004(1987). 

229. Cue* could also be identified by beneficiary complaints or by referrals from Tiscal iniermedianes. HCFA 
regional ofrices, or state medical boards. 

230. In the median PRO surveyed by the GAG recently, only 30% of the cases were speciaity-matched for the 
doctor's review. GAG Meocabe PROs. supra note 175. at 21. 



MEDICARE PEER REVIEW 1 1 1 

other than the initial reviewer. In one-fourth of the PROs I interviewed, however, the 
record and the physician's or provider's response go directly to a quality assurance 
committee. In half of the PROs I interviewed, the matched specialist reviewer refers 
problem cases on to a quality assurance committee, often at a regional level. In two 
of the remaining PROs. the matched specialist sends out a second letter noting that 
the problem has not been cleared up and requesting more information. This letter may 
be sent to hospitals at which the attending physician has staff privileges as well as to 
the doctor. Again, during this further review, cases may be classified using a grid. 
with serious cases being pursued aggressively and less serious cases being noted for 
future profile review. 

In three-fourths of the PROs I interviewed, serious or repeated problems 
eventually end up in a quality assurance committee. In about one-third of the PROs. 
this committee meets with the physician. These committees vary in size from three 
to twelve members, often meet at a regional level, and are usually composed of an 
assortment of specialists. In two of the PROs I interviewed, there are two layers of 
committee review before a case can be referred for sanctions. The quality assurance 
committee normally imposes a corrective action plan on the doctor and then monitors 
compliance over a period of time."' Some PROs monitor corrective action plans for 
a set period of time (thirty, sixty, or ninety days). Other PROs (recognizing that a 
doctor can simply avoid problem cases until a time-limited corrective action plan 
expires) monitor corrective action plans for a set number of specific kinds of cases. ^^^ 

If a problem persists through all of these layers of review, it enters the sanction 
process, which is often handled by a different committee. As is clear by now, the 
number of steps in the review process between the initial identification of a problem 
and a final decision to pursue sanctions varies considerably from PRO to PRO 
(though many of the PROs can expedite their normal processes in egregious cases). 
An AMA study found that the number of physician reviewers who review a case 
before a sanction letter is sent varied from three to thirty-five, with a mean of ten and 
a median of seven."' My impression is that the PROs with the most expedited 
procedures have traditionally been the most active in sanctioning physicians.-'-* 

The Third Scope of Work imposes a quality review process that may streamline 
this process in some PROs and complicate it in others. Under this process, the PRO 
has sixty days from the date of receipt of fiscal intermediary sampling data to obtain 
records, run them through criteria screens, and have problems assessed by a physician 
advisor. 235 Identified problems must then be reviewed within thirty days by a 

231. See GIG Sanction Rekirt. supra note 5. at 8 (describing this process further). The GIG Quality Rlport. 
supra note 167, at 18, noted that corrective action was relatively unconunon, but my impression in speaking wiih the 
PROs >*as (hat it was increasingly conunon. The AMA PRO Executive Survey, supra note 136. Table 14, found that the 
PROs the AMA interviewed had imposed from to 98 corrective action plans, with a mean of 15 6. A recent GAG study 
includes more information on corrective action plans. See generally GAG Medicare PROs, supra note 175. 

232. For e.xample. if a corrective action plan applies for 50 cases, the doctor will be monitored for 50 cases, whether 
it takes him 1 month or 24 to handle this volume. 

233 AMA PRG E.XEamvE Survey, supra note 136. Table II. 

234. The GIG Sanction Report, supra note 6, at 14. agrees with this assessment. The GIG Quauty Report, supra 
note 167, at 18. notes that, conversely, the PROs with the most complicated procedures sanction the least. 

235. Third Scope of Work, supra note 3, § V(B), at 6. 



112 TIMOTHY STOLZFUS JOST 

specialist or quality assurance committee.-'* When problems are not resolved, the 
provider must be given an opportunity to discuss the case, which consumes another 
thirty days.^^' Finally, the PRO has 15 additional days to confirm or not confirm the 
quality problem and to send final notice to the attending physician or provider, for a 
total of 135 days."* Within this 135-day maximum. PROs may adjust the time frame, 
as long as the attending physician or provider is given 30 days to discuss the 
problem."' 

The initial reviewer in the Third Scope of Work quality assurance process must 
classify problems into one of three categories depending on the severity of the 
problem. The most trivial, severity level I. problems are noted but not acted upon 
unless a pattern emerges. 2'"' Severity level II and III problems are to be profiled on 
a quaneriy basis. 2"' Appropriate interventions are to be pursued by the PRO, 
depending on the weighted score assigned to confirmed problems. ^'•^ These might 
include notification, education, intensified review, coordination with licensing and 
accreditation bodies, and, ultimately, sanctions. ^^^ Though cases appropriate for 
sanctions will normally be identified through the quality assurance process, the 
sanction process is independent, and sanctions may be initiated in appropriate cases 
even though not required by the quality assurance process. 2*** 

Once a problem is finally confirmed and a case enters the sanction process, it is 
governed by 42 C.F.R. part 1004. 2*' If the case involves a substantial number of 
substantial violations, the PRO begins by sending a sanction notice to the provider or 
physician and requesting additional information or a meeting to discuss the problem 
within twenty days of the receipt of the notice.^** For a gross and flagrant violation, 
or the case of a substantial number of substantial violations not cleared up through the 
discussions following this initial notice, a second notice is sent out requesting 
additional information or an opportunity to meet within thirty days.^*^ Both notices 
must clearly explain the problem identified, the obligation violated, the facts relied 
on, the potential consequences of a sanction, and the importance of the sanction 
meeting. 2"** At the thirty-day notice meeting, the physician or provider may have an 
attorney present who may make opening and closing remarks, ask clarifying 
questions, and assist in presenting expert witnesses."' Physicians who are biased 

236. Id. 
237 Id. 

238. Id. 

239. Id. 

240. Id § V(D). at 7. 

241. 1.1. § V(n. at 8. 

242. Id. §§ V(F>-(G). at 8-13. 

243. Id. § V(G). at 10-12. The most common intervention is a notiHcation letter, used 6 times per 1000 physicians 
per quarter in 1987-88. PROs consider sanctions to be the nxMt effective intervention for dealing with quality problems. 
GAO Medicare PROs. supra note 175. at 25-27 

244 Third Scoft of Work, supra note 3. § V(G)<4). at 13 

245. A description of this process is included in the OIG Sanction Retort, supra note 6. 

246 42C.fr § 1004 40(1987) 

247 Id. 

24«. Id. §5 1004.40. .50; U.S. Depi of Health A Human Servs. News Release. Changes m PRO Sanciion FnKtss. 
.May 13. 1987. 

249. PRO Mam'ai.. supra note S3. § 6025(B). 



MEDICARE PEER REVIEW 1 13 

against or in direct competition with the accused physician or provider, or who were 
responsible for the findings presented to the sanction committee, may not vote on the 
sanction recommendation."^ 

If the PRO decides to recommend that the OIG sanction a physician or provider, 
it sends its determination to the OIG. The physician or provider may send additional 
information within thirty days to the OIG.-" The PRO report to the OIG must 
thoroughly explain the basis for the recommended sanction. It must also recommend 
an appropriate period of exclusion or amount of monetary fine,"^ considering the 
type and seventy of the offense involved, the deterrent value of a sanction, the 
previous record of the sanctioned individual or entity, the availability of alternative 
services in the community, and other relevant factors."' The OIG must then decide, 
considering these factors, whether or not to impose a sanction and also whether the 
provider is unwilling or unable substantially to comply with these obligations. ^5"* The 
sanction is effective fifteen days from the date the physician or provider receives the 
notice from the OIG."' The OIG also provides notification of the sanction to the 
public (through a notice in a local newspaper), state Medicaid fraud control units and 
state licensing bodies, hospitals and other facilities at which the physician has 
privileges, medical societies, and medical carriers, intermediaries, and HMOs."^ 
The sanctioned provider or physician may appeal to an ALJ and ultimately obtain 
Judicial review,"'' but this may be done only after the sanction has been 
implemented."* 

D. Major Issues in the PRO Sanction Process 

1. Does the Constitution Require a Pre-exclusion Hearing? 

To this point, this Article has described the procedures afforded providers and 
physicians in the sanction process by the PRO statute, regulations, and PRO Manual. 
In a number of cases, sanctioned physicians have argued that these procedures are 
insufficient to meet the demands of the due process clause of the fifth amendment. In 
particular, they have argued that the Constitution guarantees them the right to an AU 
hearing before they are terminated from the Medicare program."' 

An analysis of the rights that should be afforded physicians and providers must 
begin with a consideration of whether their participation in the Medicare program is 
protected by the fifth amendment. As interpreted by the Supreme Court, the fifth and 



:50. Id. § 6025(CK2). 

:5I. 42C.fr. § 1004.60(1987). 

252. Id. § 1004 70. 

253. Id. §§ 1004.70. 80. 
254 Id. § 1004.90. 

255. Id. § 1004.100. 

256. Id. 

257. Id. § 1004.130. 

258. Except for rural physicians: see supra (exi accompanying notes 181-84. 

259. Ooyle v Bowen. 848 F 2d 296 (1st Cir. 1988); Cassim v. Bowen. 824 F.2d 791 (9ih Cir. 1987). Varandani 
V Bowen. 824 F.2d 307 (4th Cir. 1987); Lavapies v. Bowen. 687 F. Supp. 1 193 (S.D. Ohio 1988); Ferrer v Bowen, 
No. C86-3523 (N.D. Ohio filed July 6. 1987); Papcndick v. Bowen. 658 F. Supp. 1425 (W.D. Wis. 1987), 



1 14 TIMOTHY STOLZFUS JOST 

fourteenth amendments do not protect all expectations; they protect only life, liberty, 
and property interests. -<* While it certamly may be argued that a Medicare 
beneficiary has a property interest in contmued receipt of Medicare benefits, it is 
harder to argue that a provider or physician has a property nght in a continued 
contractual relationship with the government to provide services to Medicare 
beneficiaries. Some courts have noted, therefore, that providers, who are not the 
intended beneficiaries of the program, have no nghts beyond those provided in their 
contracts.^*' Other courts have recognized, on the other hand, the dependence of 
providers and physicians on Medicare and their expectation that they will not be 
termmated from Medicare without cause. On this basis, they have found a property 
right in continued program participation. ^<^2 Still other courts have held that the 
reputational damage and injury to practice caused to a provider terminated from the 
Medicare program implicates a liberty interest.-" Most of the cases considering PRO 
sanctions have been willing to assume the existence of a property or liberty interest 
and move on to the next question: What process is due?^** 

The question of whether sanctioned doctors are entitled to a pre-exclusion 
hearing must be answered under the Mathews v. Eldridge calculus by balancing the 
doctor's interest in greater procedural protection, the government's interest in 
expedited procedures, and the risk that the lack of a pretermination hearing will pro- 
duce an erroneous deprivation. 2" Applying this calculus, the courts have uniformly 
rejected physicians' claims for a pre-exclusion hearing. 2** First, the courts have 
tended to minimize the interest of the sanctioned physician, noting that the doctor will 
continue to be able to serve his non-Medicare patients, that he may even continue to 
care for Medicare patients without compensation and claim compensation later when 
vindicated, and that a successful conclusion of a post-termination hearing will restore 
his reputation. ^<>^ Second; they have stressed the compelling importance of expedited 



260 Board of Regents v. Roth. 408 U.S. S64. 569. S76 (1972). 

261. See Koetpel v. Heckler. 797 F 2d 858, 863-64 (10th Cir. 1986) (exclusion of doctor from Medicare 
reimbunentent for 10 years); Genatrics. Inc. v Hams. 640 F 2d 262. 264-65 (lOlh Cir). cen. denied. 454 U.S. 832 
( 1981 ) (lermiiuUKNi of Medicaid funding for nursing home); Green v. Cashman. 605 F. 2d 945 (6th Cir. 1979) (termination 
of Medicaid payments for nursing home). 

262. See Ritter v. Cohen. 797 F.2d 119 (3rd Cir. 1986) (termination of physician's participation in Medicaid 
program); Hathaway v. Mathews. 546 F.2d 227, 230 (7th Cir. 1976) (lermittation of Medicaid payments lo nursing 
home); Case v. Weinberger, 523 F.2d 602. 606 (2d Cir. 1975) (removal of Medicare patients from nursing honK); 
Lavaptes. 687 F. Supp. 1 193 (physician sanctioned by PRO); Papendick, 658 F. Supp. at 1432 (physician sanctioned by 
PRO) 

263. r^verpel ''97 F :d at 865-60; iMvup'S. 6S7 F Supp at 1201; Ferrer v B&wcn. No. C86-3523. slip o\, jI 
8 (N D. Ohio filed July 6, 1987). It is ironic that under the reasoning ot Koeipel a physician or provider terminated for 
no reason at all would not have a right to due process, since he has no property nght in participation in the Medicare 
program. A physician terminated, however, for gross and flagrant violations of professional standards of care is entitled 
to full due process protection, because his reputation has been impugned. 

264. Doylev. Bowen. 848F.2d296, 302(lstCir. 1988); Cassim v Bowen. 824 F 2d 791 . 796-97 (9th Cir. 1987); 
Vaiandani v. Bowen. 824 F 2d 307. 310 (4th Cir. 1987). 

265 Mathews v Eldridge. 424 U.S. 319. 335 (1976). 

266 Thorbus v Bowen. 848 F.2d 901 (8th Cir. 1988); Doyle. 848 F.2d at 302; Cassim. 824 F.2d at 797-98; 
Varandant. 824 F 2d at 310-1 1. Kotrpel. 797 F 2d at 868-69; Ritter. 797 F 2d at 123-24; Lavapies. 687 F. Supp. at 
1203;/'aprfu/ir*. 658F. Supp. at 143 1; County of San Diego v. Bowen. 631 F Supp. 947. 953 (S.D. Cal. 1986); Kwoun 
V. Schweiker. 528 F Supp. 1004. 1007 (E.D. Mo. 1981). 

267. Cassim. 824 F.2d at 797; Riner. 791 F.2d at 123; Ferrer, slip op. at 10; Papendick. 658 P. Supp. at 1431. 



MEDICARE PEER REVIEW 1 15 

proceedings for protecting Medicare beneficiaries from seriously deficient 
physicians.*''* Finally, the courts have noted that the multiple levels of review 
afforded physicians before the OIG issues a sanction lower the risk of error to an 
acceptable level.-'"'* Summing these considerations, the courts have uniformly held 
that even the procedures provided to physicians before the May 1987 compromise 
afforded sufficient notice and an opportunity to respond to charges to accord with the 
requirements of due process.-^" 

One can certainly argue with the weight given to the relevant considerations in 
these cases. In particular, they tend to trivialize the devastating impact of exclusion 
on physicians. Nevertheless, the unanimity of their results clearly establishes the 
constitutionality of exclusions based on PRO recommendations prior to an adversarial 
hearing before an ALJ. This does not mean, of course, that a pre-exclusion hearing 
might not make sense. This issue will be considered later. First, however, the 
procedures that should be followed in the PRO and ALJ process will be further 
considered. 

2. Notice 

Though the Constitution does not require a pre-exclusion hearing, it does require 
that a provider or physician receive notice of the grounds on which the sanction is 
based. 2^' The PRO regulations and manual set out fairly extensive and specific 
requirements for notices to be sent to the provider before the thirty-day review in a 
gross and flagrant case, before the twenty-day and thirty-day reviews in the case of 
a substantial number of substantial violations, and, finally, before OIG review in both 
situations. 272 jhe doctor is to be informed of the obligation violated, the basis for the 
PRO determination, the sanction the PRO will recommend, and the procedural rights 
the doctor is entitled to in the review process. 

Despite these specific requirements, providers have encountered problems with 
notice in the PRO sanctions process. First, there are difficulties that seem to stem 
from the unfamiliarity of the doctors running the PROs with legal process. In one 
PRO sanction decision, for example, the ALI faulted the PRO for providing the 
sanctioned doctor only with vague "issues of concern" rather than the clear and 
specific charges required by the regulations. "3 In other cases, PROs have raised new 
issues not covered by the original notice at a sanction meeting. ^'-^ The May 1987 
compromise requires PROs to use model letters that will more clearly give notice to 



268. Doyle. 848 F.2d al 302; Cassim. 824 F.2d al 799, Papendick. 658 F. Supp. al 1431-32. 

269. Cassim. 824 F.2d at 797-98: Papendick. 658 F Supp. at 1431. 

270. Doyle. 848 F 2d at 302; Varandani. 824 F.2d al 313; Koerpel. 797 F 2d at 867-68 (citing Cleveland Bd. of 
Educ V l^oudermill. 470 US. 532 (1985)). 

271. Cleveland Bd. of Educ. v. Loudermill. 470 US. 532. 542 (1985). 

272. 42 C.F R. §§ 1004.40-70; PRO Manual, supra note 53. § 6025. 

273. In re Apakupakul. No. HIX-000-34-7009. at 32-33 (June 1. 1987). 

274. Approxltnately 40% of the sanction rccommendauons rejected by the OIG in 1987 involved instances of the 
PROs failing to follow regulatory piocedures. including presumably failing to give proper notices. OIG Sanction Report, 
supra note 6. at 18. 



1 16 TIMOTHY STOLZFUS JOST 

providers threatened with sanctions. Manual Transmittal 15 includes such forms and 
will, it is hoped, alleviate this notice problem. 

The second notice issue is more narrow in focus. As stated earlier, the statute 
permits a provider or physician to be sanctioned only if it "has demonstrated an 
unwillingness or a lack of ability substantially to comply" with program 
obligations."' The regulations place the obligation to determine this fact on the OIG, 
after it considers the recommendation of the PRO."*" At no point, however, do the 
regulations require that the provider or physician be given notice as to the basis for 
this determination or an opportunity to respond to it. In Lavapies v. Bowen,^'''' the 
OIG was enjoined from imposing an exclusion until the question of willingness and 
ability was raised and discussed. If this requirement continues to be part of the law, 
the regulations should be amended to assure that the PRO gives the provider or 
practitioner notice of the basis of its unable or unwilling recommendation and an 
opportunity to respond. 

3. Confidentiality and Disclosure Issues 

The PRO confidentiality and disclosure statute and regulations have several 
provisions pertinent to the hearing process. First, 42 U.S.C. section I320c-9 imposes 
strict limitations on the disclosure of patient information. These limitations have been 
relied on to impede the access of expert witnesses called by sanctioned physicians and 
providers to information necessary for analyzing the patient care at issue in the 
sanction proceeding. Manual Transmittal 15, section 6025(C)(1), clarifies that such 
information should promptly be made available to expert witnesses and may solve 
this problem. 

A larger issue is raised by 42 C.F.R. section 476.139(a), which prohibits 
disclosure of PRO deliberations except to the OIG and HCFA. While this regulation, 
by itself, might seem to leave providers and practitioners in the dark as to why a PRO 
has sanctioned them, section 476.139(b) requires PROs involved in administrative 
hearings to give reasons for their decisions, including the detailed facts, findings, and 
conclusions that support their determinations. These rules are consistent with general 
administrative law requiring that decisionmakers issue findings of fact and conclu- 
sions of law, but prohibiting inquiry into the deliberative process to assure frank and 
free discussion. ^^^ They should not cause providers undue hardship. 

Even more significant to providers or physicians in the PRO sanction proceed- 
ings is the question of whether rhey can discover the identity of the reviewers who 
initiated or participated in the review process. Section 476.101(b) protects the 
identity of PRO reviewers as confidential. This provision is supported by the PROs, 
which believe that their always lean supply of qualified physician reviewers might dry 



275 42 use. § 1320c-5(b)(l982). 

276. 42C.F R. §§ 1004 90<<J)(7). 70(c)(4) (1987). PRO Manual, .(upra note 53. §6025. stales ihat ihe physician 
or provider may address ihe issue of ability and willingness al the PRO sanction meeting, but does not require the PRO 
to noufy the provider or practitioner of the evidence on which this determination will be made. 

277. 687 F. Supp. 1 193. 1201-02 (S.D Ohio 1988). 

278. United Suies v Morgan. 313 US. 409. 422 (1941). 



MEDICARE PEER REVIEW 1 17 

up if reviewers became subject to the threat of litigation and to other forms ot 
retaliation that might occur were their identity to become known. These fears are not 
irrational, as 42 U.S.C. section 1320c-6 only offers physician reviewers qualified 
legal immunity for their review activities. ^^ Moreover, specialist reviewers may 
legitimately fear that they will lose referrals if it becomes known that they are 
responsible for sanctioning other physicians. 

Sanctioned physicians and their representatives, on the other hand, complam 
that this provision depnves them of the opportunity to confront their accusers, a basic 
nght in Anglo-American junsprudence. Many of them believe that once a reviewing 
physician decides that an attending physician's practice has been deficient, subse 
quent reviewers tend to accept the initial reviewer's analysis uncritically. They are 
concerned that if the initial reviewer cannot be questioned, it will be difficult to 
convince the sanction committee that the initial reviewer's reasoning was flawed. 
Moreover, the regulation conceals from sanctioned physicians not only the initial 
reviewer's identity, but also his credentials. Denied access to these credentials. 
providers or physicians cannot determine whether the initial reviewer was competent 
to review the particular case. 

These interests can largely be accommodated. Providers and physicians in the 
sanction process should be apprised of the identity and credentials of reviewers who 
present evidence against them before sanction review committees and before ALJs. 
The identity of the much more numerous physician advisors and specialist reviewers 
who identify cases involving inappropriate care for referral to the sanction process 
should not be revealed, however, as long as their opinions are not relied on 
exclusively by those making the final PRO sanction decision. The credentials of these 
reviewers — for example, whether or not they are board-certified and in what 
specialty, the nature of their practice, or how long they have been in practice — should 
be revealed to the sanctioned physician or provider, if this can be done without 
revealing the reviewer's identity. 

A final disclosure issue involves the method by which beneficiaries are made 
aware of PRO exclusions. Currently, the OIG is required to publish in a newspaper 
of general circulation a notice identifying sanctioned providers. -«» The May 1987 
agreement committed HCFA and the OIG to promulgate a regulation allowing 
physicians to have the option of notifying their own Medicare patients of their 
exclusion. This option has not yet been implemented. While allowing physicians to 
notify their own patients directly may be more protective of their dignity, and may 
even make it nr.ore certain that beneficiaries will be nonfied of the sanction, it is 
difficult to understand how this provision will be enforced. The temptation would 
surely be great for a sanctioned physician to provide his or her patients with 
something less than the full truth about the exclusion or even not to send the notice 



279. They mu$i exercise "due care" to receive immuiiKy {but see Kwoun v Southeast Mo. PSRO. 811 F 2d 401 
(8th Cir. 1987). cert, denied, 108 S. Ct. 1994 ( 1988) (rioding PSRO employees absolutely immune from suit)) 

280. 42C.F.R. § 1004.100(d) (1987). 



118 TIMOTHY STOLZFUS JOST 

at all. It IS important in implementirig this provision for HHS to assure that Medicare 
beneficianes do in fact receive a prompt and accurate notice of exclusion.-"' 

4. Burden of Proof and Wetghi of Evidence Issues 

The PRO statute, regulations, and Manual do not clearly identify who bears the 
burden of proof in establishing the appropnateness of a sanction at the PRO or ALI 
level. The APA provides that "(e)xcept as otherwise provided by statute, the 
proponent of a rule or order has the burden of proof. "-"^ This would seem to put the 
burden on the PRO or the OIG seeking a sanction to prove up its case before an ALI. 
On the other hand, the sanction statute places the obligation on providers and 
physicians to assure that services provided Medicare beneficiaries are "supported by 
evidence of medical necessity and quality in such form and fashion and at such time 
as may reasonably be required" by a reviewing PRO.^"^ This would seem to support 
the position of the OIG that the provider or physician when challenged has the burden 
of establishing that services meet the Act's requirements. ^"^ 

Several ALI opinions have considered the burden of proof issue at the ALI level. 
They note that the proceeding is de novo and that there is, therefore, no presumption 
that there has been a violation for which the sanction proposed by the AU is 
appropriate.^*' As the OIG initiates the sanction proceedings, these decisions place 
upon it the burden of offering evidence to establish the statutory requirements for 
exclusion from the program.^** Once the OIG has presented this evidence, however, 
the ALI opinions split. Two ALI opinions place on the respondent the ultimate 
burden of proof that the sanction was factually unsupported and legally unjustified.^*' 
A third seems to say that the burden remains on the OIG to prove its case. 2** A fourth 
opines that the burden is "shared. "2*' 

Because physician and provider participation in Medicare has become so 
widespread and so essential to the practice of medicine, it is appropriate that the 
proponent of the sanction of exclusion at both the PRO and ALI level bear the burden 
of establishing that the exclusion requirements of 42 U.S.C. section l320c-5 have 
been met. This is consistent with practice in OIG civil penalty cases^'*' and with 



281. ABA Commission on Legal Problems for the Elderly & The Administkative Confwence of the UNrreo 
States. Medicare Procedures Symposium: Report and Recommenoahons 30 (1987) |hereinafier ABA/ACUS Recom- 
mendations | (suggests individualized mailings ^s a more efrective means of noifying beneHcianes than newspapii 
notice). 

282. 5 use. § 556(d) (1982). 

283. 42 U.S.C. § 1320c-5(a)(3)(1982). 

284. See Fairfax Hosp. Ass'n Inc. v Califano. 585 F 2d 602. 61 1 (4th Cir. 1978) (Medicare provider has burden 
of establishing validity of disallowed claim in administrative proceedings). 

285. In re Lifshutz. No. 000-44-7020. at 8-10 (Mar. 25. 1988); In re Polka, No. HIP-00O4X)^229. at 6 (June 17. 
1987). 

286 Ufshun. No. 00(M4-7020. at 10 

287 Polka. No. HIP-OOOflO-0229, at 6; In re Apakupakul. No. H I X-OOO- 34-7009. at 7 (June 1. 1987). 

288. Ufshun. No. OOO-U-7020. at 10. 

289. In re Santos. No. 000-54-7029. at 3 (Feb. 25. 1988). 

290. See42CFR. § 1003.114(1987). 



MEDICARE PEER REVIEW 1 19 

administrative law generally when a party is charged with an illegal or improper 
act.-*" 

The question of the standard of proof is less controversial. Although some 
provider attorneys have argued that the serious consequences of sanctions call for a 
clear and convincing evidence standard. ^'^ ALJs have uniformly required only proof 
by a preponderance of the evidence. ^■'^ This is consistent with the general standard of 
proof in admmistrative law^** and is appropnate in these cases, given the balance of 
weighty considerations favoring both the protection of providers and of beneficianes. 

5. The "Unwilling or Unable" Requirement 

As noted earlier, 42 U.S.C. section 1320c-5 prohibits HHS from sanctioning 
providers or practitioners unless they have "demonstrated an unwillingness or a lack 
of ability substantially to comply" with program obligations. The failure of the OIG 
to establish this element to the satisfaction of an ALJ has been the cause of several 
OIG losses in sanction cases. 2*' Failure of PROs to establish unwillingness or 
inability has also been one of the most frequent reasons for rejection of PRO sanction 
recommendations by the OIG.^'* The OIG has noted the confusion involving the 
determination of unwillingness and inability and has recommended that Congress 
delete the requirement. ^'^ This would make the PRO legislation consistent with 
section 1 128(b) of the Social Security Act, which allows the OIG to exclude poor 
quality providers without showing unwillingness or inability. 2'* It would also restore 
the law existing under the PSRO program, which permitted proof of the underlying 
violations of standards to establish unwillingness or inability to comply. ^^^ 

It is difficult to prove that providers or physicians are unwilling or unable to 
comply with program obligations. Most doctors faced with sanctions will enthusias- 
tically express their willingness to comply, and many ALJs believe that if a doctor is 
licensed he should be able to comply. In one case, for example, the ALJ concluded 

291. 12 A. C).S. Public Administrative Law and Procedure i 128(1983) 

292. See Adams & Krebs-Maricrick. Defending Physicians Before the Peer Review Organization: Practical 
Strategies and Tactics in Sanction Proceedings, in National Health Lawyejis Association. Utilization, MANAOEMthrr. 
PROs AND Quality Assurance: The Legal Pitfalls 21 (1987). 

293. In re Lifshutz. No. 000-44-7020. at 10 (Mar. 25. 1988); In re Pollca. No. HIP-000-00-0229, at 6 (June 17. 
1987): In re Apakupakul. No. HIX-OOO- 34-7009. at 23 (June I. 1987). 

294 See Steadman v. SEC. 450 U.S. 91 (1981); Bieeden v. Weinberger. 493 F 2d 1002. 1005 (4th Cir. 1974); 
Whaley v Gardner. 374 F 2d 9. 10 (8th Cir. 1%7). 

295. Lifshutz. No. 000-44-7020. at 18-20; /n re Hill. No. HIX-000-64-7015. at 5-6 (Nov. 16, 1987); In re 
Rod«b.-ugh. No PS OOC-74-7002. at 4-5 (.'une 9. 1987) 

296. OIG Sanction Rekwt. supra note 6, ;.t 18. 

297. Id. at 16. 21. 

298. 42 us C A. § 1320a-7(b)(6)(B) (West 1983 & Supp. 1988). Section 1 128 allows the OIG to exclude providers 
or physicians from Medicare who have been convicted of various crimes, lost their licenses or been excluded from other 
governmental programs, committed fraud, or refused to cooperate with various program requirements. It also permits 
exclusion of individuals or entities who provide items or services "substantially in excess of the needs of . . patients, 
or of a quality which fails to meet professionally recognized standards." This provision is used relatively infrequently, 
principally to sanction physicians for errors committed in practice settings not supervised by the PROs. such as in their 
private offices. 

299 Social Security Amendments of 1972. Pub. L. No. 92-603. § I l60(bKI). 86 Stat. 1548. 1683 (codified at 42 
use. § 1320c- 5(b)(1)). amended by Tax Equity and Fiscal Responsibility Act of 1982. Pub. L. No. 97-248, % Stat. 
324. 388. 



120 TIMOTHY STOLZFUS JOST 

that in several instances the services provided by the sanctioned doctor had grossly 
and flagrantly violated professional standards. Noting, however, that the doctor said 
he was now willing to comply, and seemed to be bnght and well-educated, the AU 
threw out the sanction recommendation.'** In another case, the AU found that the 
PRO must attempt a corrective action plan before it can establish unwillingness or 
inability. '0' On the other hand, ALIs have excluded doctors when they felt that the 
doctor lacked a basic understanding of medical symptoms and procedures.'"^ 

At the PRO level, the requirement has generally resulted in PROs imposing 
lengthy corrective action plans before considering a sanction. Most of the PROs with 
which I spoke believe that in most cases unwillingness and inability can only be 
demonstrated by the failure of a corrective action plan. There is much to be said for 
this approach, as correction through education is the primary focus of the PRO 
program. It is unfortunate, however, if this approach results in incompetent 
physicians being kept in practice until they injure a sufficiently large quota of 
patients. 

On balance, the "unwilling or unable" requirement should be repealed. If it is 
retained, however, HCFA should issue regulations making it clear that unwillingness 
and inability can be proved by establishing uncooperativeness. lack of basic 
knowledge or skills, impairment, or extreme incompetence over a period of time in 
the past, as well as through showing failure to comply with a corrective action plan. 

It is especially important that monetary penalties be imposed for deterrence 
purposes on physicians and providers guilty of gross or repeated violations regardless 
of their willingness or ability to be good in the future. One plausible interpretation of 
42 U.S.C. section 1320c-5 is that the "unwilling or unable" to comply requirement 
of subsection (b)(1) does not apply to the monetary penalty provision, subsection 
(b)(3), and one ALJ has read the statute to say this.^' Under this reading of the Act, 
further legislation is not necessary to reach this result. 

6. Money Penalties 

Section 1320c-5(b)(3) allows HHS, in lieu of exclusion, to require a practitioner 
or provider who has provided improper or unnecessary services to pay a monetary 
penalty of an amount not in excess of the actual or estimated cost of those services. 
This provision provides in theory a useful, less drastic, alternative to exclusion. 
Alternative sanctions to program exclusion that can be applied flexibly to address 
particular problems have proven very useful in other regulatory schemes, sucn as 
nursing home licensure,'** and should be available to the PROs. 

300 In re Hill. No. HIX-000-64-7015, al 5-6 (Nov. 16. 1987); see alio In re Rodabaugh. No. PS 000-74-7002. 
at 4 (June 9. 1987) (willingness of doctor to engage in continuing education and lo monitor patients enough to show 
willingness and ability). 

301. In re Lifshutz. No 00(M4-7020. at 19 (Mar. 25. 1988). 

302. See In re Betty, No. HIX-000-64-7003. at 2 (Sept. 30. 1987); In re Rivero. No. HIS-000-64-7010. at 24-25 
(Sept. 29. 1987). 

303. In re Santos. No. 000-54-7029. al 22 (Feb. 25. 1988). 

30* See Instttvte of MEOictNE. Imhiovino the CJuauty of Cars in Nursing Homes 162-68 (1986). See also 1 
C.F R. § 305.79-3 (1987) (ACUS recommendations regarding civil penalties). 



MEDICARE PEER REVIEW 121 

Unfortunately, the ALJs enforcing the penalty provision have read it very 
narrowly to limit monetary penalties to the amount of isolated charges stnctly 
applicable to a particular medical procedure determmed to be unnecessary or of poor 
quality. '<'* The result in one case was a penalty of $65.44. ^o** Obviously penalties of 
this size do not justify the cost of a sanction process, and the OIG has effectively 
ceased using the monetary penalty authonty or accepting PRO recommendations for 
penalties. ^^ 

The PRO legislation ought to be amended to allow the OIG to impose monetary 
penalties of up to $10,000.^* This would allow the OIG to impose a sanction that 
providers and physicians would take seriously but would be short of exclusion from 
the program. As stated above, this sanction should be for past violations of physician 
and provider obligations and be aimed at deterring future violations. It should not, 
therefore, be subject to the "unwilling or unable" to comply requirement. 

7. Bias 

One of the most fundamental rights afforded by due process to a person subject 
to an administrative adjudication is the right to a hearing before an impartial 
tribunal. ^^'^ One of the most persistent complaints made against the PROs by 
physicians and providers is that their hearing procedures are inherently biased against 
providers and practitioners accused of sanctionable conduct. This charge takes three 
forms. 

First, review by one's peers can also mean review by one's competitors. There 
has been a general concern that reviewers who participate in the quality assurance and 
sanction process may be competitors of the sanctioned physician, eager to get him or 
her out of the way to expand their own practice. Conversely, there is the concern that 
PRO reviewers may be friends and associates of those whose care they review, biased 
against the public whom they should be protecting. These concerns are addressed to 
a minimal extent by the PRO regulations, which prohibit persons from reviewing 
cases if they have participated in treating the beneficiary, are a member of the 
beneficiary's family, or have a management or ownership interest in the facility at 
which services were furnished to the beneficiary. "» The PRO Manual further 
requires PROs to assure that members of any hearing panel making a final sanction 
decision are not affected by personal bias or direct economic competition with the 
accused provider or practitioner. 

In practice, most PROs go beyond these requirements to avoid this sort of bias. 
AH of the PROs with which I spoke refuse to allow physicians to review cases from 
hospitals or HMOs at which they have staff privileges to guard against both prejudice 



305. See Sanios. No. 000-54-7029. al 19-20; In re Polka. No. HlP-OOaOO-0229. al 20 (June 17. 1987) 

306. OIG QuAunr Retort, supra no«e 167. at 9. 

307. See OmcE of Inspectok GbNEiiAL. PRO Technical Information Memorandum No. 2, at 3 (July 24. 1987) 
(rines only cost effective when an improper pattern of care involving substantial reimbursement is involved). 

308. See ABA/ACUS Recommendations, supra note 281. at 30; OIG Sanction Reiort. supra note 6. at iii. 

309. Withrow v. Larkin. 421 U.S. 35. 46 (1975). 

310. 42 C.F R. § 466.98(d) (1987). 



122 TIMOTHY STOLZFUS JOST 

and favoritism. One even had a computer system to assure that this would not happen. 
Several went further, refusmg to allow reviewers to review cases from their own 
community or from communities from which they received referrals. This policy was 
particularly prevalent in rural areas. Other PROs do not allow doctors on statewide 
sanction committees to vote on cases from their region and allow the provider or 
practitioner subject to sanction proceedings to object to committee members who are 
suspected of bias. These protections seem adequate to the job and have been accepted 
by the one case that has directly addressed this bias question.'" 

A second concern is that the PROs inappropriately combine the prosecutorial 
and adjudicatory function — that is, reviewers who recommend sanctions are permit- 
ted to participate in the decisionmaking process. It is generally inappropriate for an 
administrative officer to participate in both the prosecution and decision of a case.^'^ 
though the Supreme Court has rejected calls for strict separation of investigative and 
adjudicatory functions.''^ The one PRO case in which the issue was raised rejected 
a strict separation requirement.' '* On the other hand, ALJs have rejected the 
testimony of PRO experts who participated in a sanction investigation as not 
credible.'" 

Manual Transmittal 15, growing out of the May 1987 compromise, provides 
that a physician who "was solely or primarily responsible for making medical 
judgments and developing the record and initial findings to be used at the discussion 
shall not vote on the PRO's final determination about whether or not to recommend 
a sanction to the OIG.""*This provision does not apply, however, to physicians who 
summarize the views of others to assemble a record and findings for the sanction 
meeting. It provides a useful compromise between the need to protect the rights of a 
physician or provider threatened with a sanction and the need of the sanction 
committee to have available someone with knowledge of the case. My impression 
from interviews with the PROs is that this provision has not yet been implemented by 
all of the PROs. It should be. 

Finally, the most serious charge of impropriety against the PROs is that they are 
inherently biased against providers because renewal of their contracts depends on 
their meeting a "quota" of sanctioned providers."^ The PROs operate pursuant to 
contracts with HHS, and their multimillion dollar contracts depend on meeting 
contract objectives. Provider attorneys have frequently claimed that favorable 
evaluations of the PROs by HHS, and thus renewal of their contracts, depend heavily 
on the number of sanctions imposed by the PROs. They also claim that OIG officials 



311. Uvapics V. Bowen. 687 F. Supp. 1193. 1203-04 (SO. Ohio 1988). 

312. Amcncan Cyanamid Co. v FTC. 363 F.2d 757 (6th Cir. 1966). 
313 Wiihrow. 421 U.S. 35. 

314. Doyle v. Bowen. 660 F, Supp. 1484. 1488 (D. Me. 1987). rev d on other grounds. 848 F.2d 296, 302 (Isl 
Cir. 1988). 

315. In re Polka. No. HIP-000-00-0229. at 17 (June 17. 1987); In re Apakupakul. No. HIX-000-34-7009. at 32. 
35 (June I. 1987) 

316. PRO Manual, supra note 53. § 6025(CK5). 

317. See Carlova. supra note 219. at 62. 



MEDICARE PEER REVIEW 123 

are rewarded for wringing sanctions out of the PROs.^'" Thus the PROs and OIG 
must find victims to sanction, whether or not those victims actually are guilty of 
sanctionable offenses.'" 

It is, of course, improper for an administrative tnbunal to hear a case in which 
it has a financial interest. '^^ Courts that have addressed this allegation, however, have 
found that the PROs are not improperly biased in this respect. Doyle v. Bowen^^^ 
noted that although contract renewals depended on meeting contract objectives, 
including sanctioning substandard care, it had never been suggested that the PROs 
should sanction physicians who were providing adequate care.'^^ Moreover. Doyle 
noted that there was no evidence that individual reviewers in the case were in any way 
influenced by the PRO's concern about imposing an adequate number of 
sanctions. '23 Similarly the court in Lavapies v. Bowen rejected the bounty system 
argument and accepted the testimony of the Ohio PRO's Vice President of Finance 
that the HHS had never imposed a quota of sanctions. '^^ 

The HHS OIG has shown great interest in PRO sanction activity, which could 
be interpreted as pressure on the PROs. Some of the PROs have engaged in highly 
questionable conduct in pursuing PRO sanctions,'^ which might be attributable to 
overcagemess. Nevertheless, PROs that have issued no sanctions have routinely had 
their contracts renewed, demonstrating that no absolute quota system is in operation. 
This is, however, a subject to which the PROs and HHS must be sensitive in the 
future. 

E. Two Alternatives for Improving the PRO Sanction Process 

1 . Transfer Sanction Authority to the Inspector General and Provide 
Pretermination Hearings 

Many of the problems in the PRO sanction process can ultimately be traced to 
the difficulty of engrafting enforcement functions onto what is basically a peer 
monitoring and education program. The tension between enforcement and educative 
functions in health care regulation has been noted elsewhere. '^6 An effective 



318. The California Medical Associacion in Greene obtained documents sumcicntly supporting these charges lo 
obuin further discovery of federal officials. See Greene v. Bowen. No. CIV S-86-0625 LKK (E.D. Cal. 1988) (order of 
Jan. 22. 1988. at 28); Brief for Plaintiff in Opposition lo Federal EVfendants' Motion to Dismiss or for Summa-^ 
ludgment ai 7-11. Greene v. Bowen. No. S-86-0625-LKK (E.D. Cal. I9«8) (filed Sep.. 14. 1987); McGra*-Hiu. 
Utiuzatton Review Report 3 (Mar. 24. 1988). 

319. One provider anorney with whom 1 spoke noted thai the absence of judicial review for contract denials, 42 
U.S.C. § I320c-2(f). made the PROs even more concerned about HHS pressure to sanction providers. 

320. Marshall v Jerrico. Inc.. 446 U.S. 238 (1980); Ward v. Village of Monroeville. 409 US. 57 (1972); Tumey 
V Ohio. 273 U.S. 510(1927). 

321. 660 F. Supp. 1484 (D. Me. 1987). revdon other grounds. 848 F.2d 296 (1st Cir. 1988). 

322. Doyle. 660 F. Supp. at 1488. 

323. Id. 

324. Lavapies v. Bowen. 687 F. Supp. 1 193. 1203-04 (S.D. Ohio 1988); see also Association of Am Physicians 
& Surgeons v. Weinberger. 395 F. Supp. 125. 139-40 (N D. 111. 1975) (rejecting the bias charge). 

323. See In re Ufshutz. No. 000-44-7020, at 23-28 (Mar. 25. 1988); Doyle. 660 F. Supp. at 1488-89. 

326. See Oay & Klein. The Regulation of Nursing Homes: A Comparative Perspective, 65 Mil3ank Mem. Fed. Q. 



124 TIMOTHY STOLZFUS JOST 

education program must work with problem physicians in a close consultative 
relationship based on (he assumption that the physician wants to improve his practices 
if he can be taught how. An effective enforcement program assumes that there are bad 
as well as good physicians, and it maintains its distance from all as it attempts to 
effectively ferret out and prosecute the bad ones. In combining these functions. PROs 
have understandably tended to err on one side or the other. Several PROs have 
become so focused on correction, and have offered so many layers of review to assure 
giving problem physicians every opportunity to straighten out. that they have yet to 
impose a sanction. Other PROs have been overly aggressive. The basic unfamilianty 
with legal due process of the physician managers of this latter group of PROs has 
exacerbated the inquisition-like atmosphere that physicians have expenenced before 
them. 

I believe that the public, physicians, and providers would be better protected if 
the sanction authority were withdrawn from the PROs and given to the OIG. Under 
this proposal PROs would retain their quality assurance functions. They would 
continue to monitor medical records and investigate complaints. They would continue 
to identify quality and utilization problems. When quality problems were identified, 
they would continue to require corrective action. This could include the intei^entions 
suggested by the Third Scope of Work: notification, education (including telephone 
discussions, suggested literature reading, continuing medical education, meetings, 
and self-education courses), intensified review, and other interventions (including 
concurrent predischarge review, second opinions or preadmission review, or referral 
to hospital infection control, tissue, or quality assurance committees). ^^^ It might also 
include other interventions used by PROs with which I spoke: requirements of 
limiting practice, consultation, preceptorships, attendance in a residency program, 
obtaining board certification, and oversight in surgery. HHS should assist in 
disseminating information among the PROs as to novel and successful approaches to 
quality interventions. ^^s 

When a PRO identified a gross and flagrant violation of physician or provider 
obligations or a failure to cooperate with or satisfactorily complete a correction plan, 
however, the PRO would refer the case to the OIG. It would not, as now. send a 
recommended sanction to the OIG, but rather it would refer the case to the OIG for 
investigation. The OIG would then, using its own medical resources (which it would 
have to create as its resources are now very limited) or borrowing experts from the 
PRO, conduct its own investigation and build its own case. Once it decided to 
proceed against a doctor for exclusion or for a monetary penalty, the provider or 
physician could appeal directly to an AU. 

Appeals of exclusions or money penalties should be heard by the ALJs attached 
to the Departmental Grant Appeals Board. These ALJs deal routinely with sanction 



303. 307-1 1 ( 1987); Josl. Enfonemem of Quality Nursing Home Care in ihe Legal System. 13 L>w, Med. and Heal™ 
Cahe 160. 162-63(1985). 

327. Thibo Scoft oe Wouk. supra no« 3. i IV(GXI>- 

328. OIG QuAiiTY Rekmct, supra ootc 167, at ii. 



MEDICARE PEER REVIEW 125 

appeals"' and are more experienced in presiding over adversary proceedings than are 
the Social Security Office of Hearings and Appeals (OHA) judges. Appeals from the 
decisions of these ALJs would be to the Departmental Grant Appeals Board."" The 
sanction, however, would be effective immediately upon the decision of the ALI. 

This system would solve a multitude of problems. First, the PROs could focus 
on what they do best — peer review. They could also greatly simplify their current 
quality review procedures, which in many PROs have become imf)ossibly labynn- 
thine. The May 1987 compromise has effectively given providers and physicians a 
full due process hearing before the PRO prior to their being sanctioned. Providers and 
physicians charged with committing a substantial number of substantial violations are 
now entitled to two hearings before the PRO. Sanctioned providers and physicians are 
entitled to a further de novo hearing before an ALI after the sanction is imposed. 
Under the OBRA '87 amendments, rural practitioners, who have been the subject of 
two-thirds of PRO sanctions to date, are also entitled to an additional preliminary 
hearing before an ALJ before their sanctions can go into effect. Many of the PROs 
interpose additional levels of review, involving up to thirty-five reviewers, before a 
case even enters the sanction process. ''' It is difficult to believe that a system this 
cumbersome and time-consuming is truly protecting the public when dangerously 
incompetent doctors are involved. 

Because a referral for investigation implicates no property or liberty interest, "^ 
no PRO hearing would be necessary prior to the initiation of such an investigation. 
F*rocedures imposed by the Third Scope of Work — an initial review by a physician 
advisor, a further review by a matched specialist or quality conrunittee before a letter 
is sent to an attending physician or provider requesting an explanation, and a final 
review of the response received before a quality problem is confirmed (presumably 
by a specialist or committee) — would still be appropriate. These procedures assure 
that the PRO is fairly certain that there is a quality problem before even raising the 
issue with a physician or provider, and they give the physician or provider a chance 
to explain before a problem is confirmed. Once a problem is confirmed, however, the 
PRO should, without further delay or waste of resources, decide to (I) continue to 
profile the physician or provider, perhaps intensifying review, if the problem were 
minor, (2) initiate a corrective action plan if the problem were more serious but 
correctable; or (3) refer the problem to the OIG if it seemed serious. 

The OIG, with its law enforcement experience and resources, could investigate 
all referrals. When the investigation substantiated problems, the OIG would seek 
exclusion or a fine. The OIG could deputize PRO physicians to serve as experts in 
investigating and presenting the case, but for this purpose they would serve as OIG, 
not PRO, experts. As the OIG's experts would be responsible for presenting evidence 



329. See 53 Fed. Reg. 25.543 ( 1988) (assigning AUs responsibility for heanng civil money penalty and exclusion 
appeals under most other HHS sanction statutes). 

330. See Civil Money PENALTiia Rtp.. Summer 1988. at 6-7 (describing the Boards AUs). 
331 AMA PRO Executive Survey, supra note 136. Table II. 

332. Cf. Setliff v. Memonal Hosp. of Shendan County. 850 F 2d 1384 (lOlh Cir. 1988) (hospital investigation of 
physician does not affect property or liberty interest). 



126 TIMOTHY STOLZFUS JOST 

in any hearings and the sanction decision would be solely based on this evidence, 
there would be no problem with maintaining the confidentiality of PRO reviewers 
who had earlier identified the problem. If a problem were substantiated by the OIG 
investigation, the provider or physician would be sent a notice, and a heanng would 
be scheduled on an expedited basis, perhaps within sixty days. The case would be 
handled by attorneys from the OIG's office, who could avoid the problems that have 
occurred when PROs have held hearings without counsel or with less expenenced 
counsel. As the PRO would be merely responsible for referring the case for an 
investigation, and not for deciding the case, the charges of bias recounted above 
would no longer be relevant. Hearings would be held at the regional offices to make 
them accessible to physicians and their attorneys. 

From the perspective of providers and physicians, this system should be superior 
because it would assure a hearing before an impartial ALJ before exclusion. Providers 
and physicians would no longer have to put up with the quasi-club meeting or the 
quasi-inquisition atmosphere that has prevailed in some PRO sanction proceedings. 
They would be dealing with hearing officers experienced in legal process and free 
from bias. Beneficiaries should welcome the proposal because it would assure that 
dangerous physicians would be dealt with swiftly and effectively and leave the PROs 
to focus on education and correction. HHS should welcome this propnjsal because it 
would give HHS more direct control over the sanction process. The OIG is already 
familiar with this kind of process through its experience under its section 1 128(b) 
authority, reaffirmed by the Medicare and Medic::id Patient and Program Protection 
Act,^" and should be comfortable with it. Most of all, the PROs should welcome this 
proposal because it would free them from their greatest headache. They would still 
have the ability to threaten a recalcitrant provider with a referral to the OIG, but 
would themselves be less the brunt of provider and physician hostility and distrust. 

It is essential to this proposal that these cases be assigned to AUs assigned to 
the Departmental Grant Appeals Board and not to the OHA Judges who currently hear 
sanction cases. The primary reason for this is that OHA ALJs are not used to the 
exf>edited time frames necessary if problem providers are to be given pre-exclusion 
hearings. ^'^ Further, OHA ALJs are not used to dealing with adversarial proceedings 
and have approached sanction cases as though they were Social Security cases, in 
which they are responsible for protecting the claimant as well as for deciding the case. 
One example of the incorporation of Title II social security disability concepts into 
sanction proceedings is the Polka case, in which the judge gave extra weight to the 
testimony of the treating physician (as is noimally done in disability cases), even 
though the treating physician was on trial. ^^ HHS attempted at one point to develop 
a trained cadre of judges within the OHA to handle sanction cases, but turnover 

333 42 U.S.C.A. § 1320a-7 (Wesi 1983 & Supp 1988). See supra noce 298. 

334 As was related earlier. AUs have not met the 4S-<lay time limit set by HHS for decisions on the pre-exclusion 
hearings required by the Hall Amendment of OBRA '87. A recent OIG report notes that final decisions in exclusion caM:s 
lake even longer. It is currently taking about 15 months from the date of a PRO sanction recommendation and 10 months 
from an OIG sanction recommendation to the conclusion of an AU hearing. OIG Sanction Report, supra note 6. at 16. 
18. 

335. In re Polka. No. HIP-00(W)0-0229. at 17 (June 17. 1987). 



MEDICARE PEER REVIEW 127 

dispersed the group, which no longer exists. DepartmentaJ Grant Appeals Board 
ALJs have experience in handling sanction cases and should take over these cases as 
well. 

Two final proposals should be considered as ancillary to this larger proposal. 
First, the OIG should be given summary suspension power for egregious cases, as is 
possessed by many state medical boards.^'* This would operate much like a 
temporary restraining order to remove a clearly dangerous doctor from practice for a 
short period of time (fifteen to thirty days) until an expedited hearing could be held. 

Second, if the proposal suggested above were adopted, the OIG would 
effectively become a national medical board to protect Medicare beneficiaries, it may 
be advisable for the OIG to have such authonty, as there is ample evidence that the 
medical boards of many states have not been effective in disciplining the 
profession. ^3' When medical boards are effective, however, it might make sense to 
have a procedure, such as app>ears in some civil rights and environmental laws, to 
permit the OIG to certify the state procedures as equivalent in effectiveness to the 
federal procedures and have the PROs refer cases to the state medical board for 
investigation rather than to the PRO. 

2. Retain Current System, but Streamline and Assure Due Process 

While there is much to commend the idea of shifting the sanction authority to the 
OIG, it is also likely to face opposition. The PROs may worry that they will lose their 
most important lever for securing cooperation from recalcitrant providers and 
physicians. Providers and physicians may be concerned with the fairness of the 
proposed process, given their suspicions regarding the current Inspector General. 
Beneficiaries may be anxious about the added delay that the proposal could add to the 
process, which already takes too long. The OIG can be expected to object that unless 
substantial additional resources were offered to it, it could not adequately handle the 
additional responsibility. While these objections are not insurmountable, they do 
counsel consideration of a more cautious approach. 

Such an approach should be adequate to address, on the one hand, the concern 
of providers and physicians that the current system does not guarantee them access to 
an undisputably impartial presanction decisionmaker and is excessively uneven in its 
results and, on the other hand, the concern of beneficiaries that the current process 
has been too much consumed with bureaucratic delay. 

First, the twenty-day notice and opportunity to submit additional information or 
discuss the problem with the PRO, provided by 42 C.F.R. section 1004.40 for cases 
involving a substantial number of substantial violations, should be abolished. It no 



336 See Ohio Rev Code Ann. i) 4731.22(D) (Anderv>n 1987); Federation of State Medical Boards, A Guide 
TO TME Essentials of a Modern Medical Practice Act 18 (1977). reprinted in H.R. 51 10. 99ih Cong.. 2d Scss. 409 
(1986) 

337 See Office of niE Inspector General. U.S. Dep't of Health & Human Servs., Medical Licensure and 
DisctPUSE .An Overview 12-15 ( 1986). reprinted in H R. 51 10. 99th Cong.. 2d Scss. 256-59 (1986) Ihereuiafler OIG 
Liclnslre & Discipline Report]; Rayack. Mediial Licensure: Social Costs and Social Benefits. 7 Lj\w & Hum. Behav. 
147. 154-57(1983) 



128 TIMOTHY STOLZFUS JOST 



doubt made sense at one lime, when many PROs still had abbreviated procedures and 
the thirty-day notice meeting provided by section 1004.50 was quite perfunctory, to 
give physicians and providers with a pattern of problems an early opportunity to 
explain or correct their behavior. The current Scope of Work provides thirty days 
notice and an opportunity to discuss problems as part of the presanction quality 
assurance process.'^* It also focuses PROs on corrective interventions rather than 
sanctions for remediable problems."' The May 7, 1987. compromise procedures 
substantially enhance the protections available to providers and physicians in the 
thirty-day notice hearing. Under these circumstances, the additional twenty-day 
opportunity for discussion under section 1004.40 seems redundant and an unneces- 
sary cause of delay. 

Second, the Third Scope of Work should be clarified to assure that the sanction 
process is initiated immediately upon the confirmation of a quality problem, if it is 
a problem for which a sanction is the appropnate intervention. Section V(B) of the 
Third Scope of Work requires PROs to issue a final notification to an attending 
physician or provider within a maximum of seventy-five days of the identification of 
a problem by a physician advisor. ^*^ During this period the problem must be reviewed 
by the initial physician advisor and a separate specialist or quality committee, and the 
physician or provider must be notified of the problem and given an opportunity to 
discuss it. The confirmed problem will usually not be sufficiently senous to warrant 
a sanction, in which case some other intervention, such as education or intensifica- 
tion, will be appropriate.'*' If the problem is sufficiently severe to warrant a sanction, 
however, the PRO should move immediately into the section 1004 50 process, with 
the thirty-day notice serving to begin the process. Further delays for further 
committee reviews or discussions cannot be Justified. 

Third, the OIG and HCFA could do more to assure greater reliability and 
uniformity in the sanction process. Training sessions should be offered, for example, 
by the OIG to lawyers representing PROs to make certain that they understand the 
procedures the PROs must follow in the sanction process and to educate them as to 
errors that conmionly result in OIG rejection or ALJ reversal. Model forms could be 
developed by HCFA to help assure that the PROs address the factors enumerated in 
sections 1004.80 and 1004.90, which must be considered by the PRO and by the OIG 
in imposing sanctions. 

Under this alternative proposal, once the PRO initiated a Medicare exclusion 
and the OIG confirmed it and gave notice to the physician or provider, an exclusion 
would go into effect within fifteen days, as provided under the current regulations.^^ 
The physician or provider could, however, within ten days request a stay of the 
exclusion, as is currently possible for rural physicians and providers under the Hall 



338. Thiro Scope of Work, supra note 3. § V(B). at 6. 
339 See generally id. § V. al 5-15. 
3J0. Id. § V(B). al 6. 

341. Id. §§ V(GM3M4). at 12-13 (which gives guidance to PROs as to the circumstances in which a sanction is 
appropnate). 

342 42C.fr. § 10O4 100(b) (1987). 



MEDICARE PEER REVIEW 129 

Amendment,^" if the stay would pose no senous threat to the patients of the 
physician or provider. The proposal would, however, establish uniform procedures 
for all, recognizing that there is no defensible reason for distinguishing between rural 
and urban physicians or providers. It would place the burden of proof on physicians 
or providers to show that their continued participation in the Medicare program would 
not pose a serious nsk to their patients. By the time the sanction proceeding reaches 
this point, the physician or provider has already had at least three opportunities to 
explain the problem on which the sanction is based — once in the quality assurance 
process, again at the formal meeting with the PRO (to which lawyers and witnesses 
can be brought), and again when the case is referred to the OIG. Both the PRO and 
OIG, considering all of the information presented by the physician or provider 
through this process, have concurred that exclusion is appropriate because either a 
gross and flagrant violation or a substantial number of substantial violations has been 
confirmed. At this f)oint it is proper that the provider or physician establish why, in 
spite of this, the exclusion should be stayed. If the provider or physician asks for such 
a stay, the effective date of the exclusion should be delayed for thirty days, during 
which the AU should consider the record and any additional information supplied by 
the physician, provider, OIG, or PRO. The ALJ should render its decision on the stay 
within thirty days. Regardless of the decision reached by the ALJ on the stay issue, 
the ALJ should reach a final decision on the merits within one year to assure that, on 
the one hand, the stay does not continue indefmitely or, on the other hand, that the 
exclusion is not stayed for an excessive period of time. 

VL Deniai^ of Payment for Substandard Care 

COBRA '85 amended the PRO legislation to require that PROs deny payment 
for services provided Medicare beneficiaries when the quality of those services does 
not meet professionally recognized standards of health care.'** The amendment 
further required that such denials should be made only "on the basis of criteria which 
are consistent with guidelines established by the Secretary.'*'^' This provision has 
proved intensely controversial, and two-and-one-half years after its enactment it has 
still not been implemented by HHS, although proposed regulations have been 
published very recently.'^ 

The Judgment of Congress that the Medicare program ought not to pay for poor 
quality care provided to Medicare beneficiaries makes a great deal of sense. It also 

343. 42 U.S.C.A. § l320c-5(bH5) (West 1983 & Supp. 1988). 

344. COBRA '85. Pub. L. No. 99-272. 1986 U.S. Code Cong. & Admin. News (100 Stal.) 83. 200 (to be codified 
at 42 use. § 1320c-3(a)(2)). 

345. OBRA '87 fuither provides that the PRO shall not disapprove payment until 20 days after the PRO has notified 
the affected provider or practitioner and afforded an opportunity for discussion and review. Pub. L. No. 100-203. 1987 
U.S. Code Cong. & Admin. News (101 Stat.) 1330. 1330-135 to -136 (to be codified at 42 US C. § 1320c-3(a)(3)). and 
prohibits physicians from charging beneficiaries for services for which payment is denied because of substandard quality. 
id. at 1330-139 (to be codified at 42 U.S.C. § I395u). 

346. See Proposed Substandard Care Regulations, supra note 61. For discussions of ihe reasons for ihe 
long-delayed issuance of the proposed regulations, see Hospitals Anxious Over Payment Denials. HosmAi^. June 20. 
1987, al 48. 53: Payment Denials Spark Questions. HosPtTALS. June 20. 1987. at 32. Paxment will be Denied for 
Substandard Care. Medical World News. Jan. 12. 1987, at 25. 26. 



130 TIMOTHY STOLZFUS JOST 

makes sense to notify beneficiaries if the PRO peer review process concludes that the 
beneficiary has been provided substandard care. Hospitals and doctors fear that the 
PRO notification of substandard care denials may lead to increased malpractice 
litigation. But if a patient has been injured by poor quality medical care, he or she has 
a legal right to sue for redress for the injury. It is difficult to understand why the PRO 
should assist in covering up the fact that such a patient has received substandard care. 

It is, of course, important for the PRO to be very sure that care provided a 
beneficiary was in fact substandard before the beneficiary is notified of this fact. The 
statute, however, assures physicians and providers that the PRO can only deny 
payment pursuant to articulated quality criteria and after notice and an opportunity for 
discussion.^' Proposed regulations further protect providers by only allowing 
payment denials when substandard quality care has resulted in "[a]n actual, 
significant, adverse effect" or "[a)n imminent danger to the health, safety, and 
well-being of the beneficiary, or places the beneficiary unnecessarily in a high-risk 
situation"^*** and, in most cases, after the case has been reviewed by a specialist in 
the area of care at issue. ^' The proposed regulations would define "actual, 
significant, adverse effects" rcstrictively to include patient management- that results 
in "(i) [u)nnecessarily prolonged treatment; (ii) [m)edical complications; (iii) 
[r]eadmission; (iv) [pjhysiological or anatomical impairment; (v) {djisability; or (vi) 
[dleath."'«» 

The delay of HHS in implementing the substandard care provision reached the 
level of unconscionability and, perhaps, illegality. In other contexts, excessive 
administrative delay has been challenged by the courts.^" If the proposal made earlier 
in this Article to transfer sanction authority to the OiC were adopted, it would 
become even more important that PROs have the power to deny payment for 
substandard care in order to assure them some ability to respond directly to poor 
quality providers and practitioners. HHS's proposed regulations, therefore, should be 
implemented expeditiously. 

Though the problems addressed by the NPRM as to how to evaluate the quality 
of medical care are largely beyond the scope of this Article, one comment will be 
made here. HHS should reconsider its proposal to notify beneficiaries only that 
payment is being denied because care was substandard and not to reveal the specific 
reason for the decision. ^'^ jhe draft rule of HHS mandates that payment only be 
denied when care is significantly deficient and results in actual significant adverse 
effects or an imminent threat to the patient. In these circumstances, the beneficiary 
deserves to be told the nature of the substandard care, both to allow the beneficiary 



347. 42 use § l320c-3{aH2)-(3) (1982 & Supp. IV 1986). 

348. Proksco Substandard Care Recui-ations. supra note 61. al 1963 (to be codified at 42 C.F R. p« 462. 
§ 466 100(b)(4Xi). (li)). 

349. W. (to be codified at 42 C.F R. pi. 462. § 466 98(aH4)) 

350. IJ. (to be codified at 42 C.F R. pt. 462. § 466. 100(b)<4Ki))- 

351. See Nider v FCC. 520 F.2d 182. 206 (DC. Cir. 1975): EnvironmentaJ [defense Fund. Inc. v. Hardin. 428 
F.2d 1093. 1099 (D.D.C. 1970) (holding that agency inaction over a prolonged penod of time can warrant judicial 
intervention). 

352. PRorasEO Substandard Cake Rtcvukvonsjupra note 61. at I960. 



MEDICARE PEER REVIEW 131 



to take necessary remedial action to deal with the threatened or actual adverse effects 
and to relieve, or at least to focus, anxiety that the beneficiary may suffer from 
receivmg a general notice that he or she has received dangerously deficient care. A 
more specific notice may result in marginally more negligence litigation, but after all. 
the PRO program exists to protect the beneficiary, not the physician. 

VII. PRO Investigations of Beneficiary Complaints 

From the beginning, PROs have based their quality review activities primarily 
on the hospital records they assemble through the sampling protocols described in 
Part II."' OBRA "86 required PROs to consider a new source of data for identifying 
health care quality problems: beneficiary complaints. 

Section 9353(c) of OBRA "86 requires that PROs: 

Conduct an appropriate review of all written complaints about the quality of services (for 
which payment may otherwise be made under title XVIII) not meeting professionally 
recognized standards of health care, if the complaint is filed with the organization by an 
individual entided to benefits for such services under such title (or a person acting on the 
individual's behalO- The organization shall inform the individual (or representative) of the 
organization's final disposition of the complaint. Before the organization concludes that the 
quality of services does not meet professionally recognized standards of health care, the 
organizadon must provide the practitioner or person concerned with reasonable notice and 
opportunity for discussion.'** 

HHS has implemented this provision through a modification in the contract and 
the Third Scope of Work and only very recently has proposed an implementing 
regulation. Under the Third Scope of Work, PROs are to investigate complaints about 
hospital inpatient or outpatient care, skilled nursing services, home health agencies, 
and ambulatory surgical centers.^'* PROs may only investigate written complaints. 
Once a complaint is received, the PRO must determine whether the complainant is a 
beneficiary or beneficiary representative, whether the services were provided in a 
Medicare-certified facility or part of a facility, and whether the services complained 
of are covered by Medicare (whether or not they were covered for this particular 
beneficiary).'" 

Once the PRO determines it has jurisdiction over the complaint, it initially 
assesses the complaint to determine if investigation is warranted. If the PRO 
determines that the complaint has merit, it must request the beneficiary's medical 
records within fifteen calendar days.'" The facility has thirty calendar days to 
provide the record.'^" The PRO review must be completed within fifteen calendar 
days thereafter if no quality problem is identified. '5* If a quality problem is found, 

353. See supra text accompanying notes 38<-40. 

354. OBRA 86. Pub. L. No. 99-509. 1986 U.S. CooeCoMO. & Admin. News ( 100 Sui.) 1874. 2047 (lo be codified 
at42U.S.C. § I320c-3(a)). 

355. Thuu) Scope of Woiuc. supra noce 3. S XVII(A). at 36-38. 

356. Id. § XVIKAKl). ai36. 
357 Id. § XVII(AK4Ka). ai 37. 

358. Id. i XVII(AH4Kb). ai 37. 

359. Id. § XVlI(AH4Xc). at 37. 



132 TIMOTHY STOLZFUS JOST 

however, the PRO must contact the physician or provider and afford another thirty 
days for discussion.^** Within five days after the review is completed, the PRO must 
respond to the beneficiary. Finally, if the PRO identifies a quality problem, it may 
initiate sanctions or otherwise proceed through its quality assurance process, and it 
must notify the regional office.'*' 

The Third Scope of Work places two restrictions on the PRO's duty to respond 
to the complaining beneficiary. First, it requires that under certain circumstances the 
response should be to someone other than the beneficiary. The Third Scope of Work 
provides that within the review time frames set out above the PRO must contact the 
attending physician of the patient whose care is the subject of the complaint fifteen 
days before disclosing information to the patient, as provided by 42 C.F.R. section 
476. 132. Section 476. 132 addresses disclosure to patients of information concerning 
themselves. It requires the PRO to seek the advice of anending practitioners regarding 
the appropriateness of releasing information to a patient. More particularly, it 
requires that, when an attending physician believes that disclosure may harm the 
patient, the PRO must disclose the information to the patient's designated represen- 
tative rather than to the patient or, when the patient is mentally, physically, or legally 
unable to designate a representative, to a person the PRO determines to be 
"responsible for the patient."'" 

Second, the Third Scope of Work differentiates between physicians and 
providers as to the appropriate PRO response to a beneficiary complaint. If the 
complaint involves a facility, the PRO must inform the beneficiary as to quality 
deficiencies it has discovered and corrective action it requires.'*' It must include 
within this notice any comments the facility made during the thirty-day discussion 
period. If the quality problem involves a physician, however, the PRO may only 
provide "(a) general response that assures the beneficiary that a thorough investiga- 
tion of his/her complaint is being conducted and that corrective action will be taken 
when a problem is found."'** 

The PROs have received few complaints through this process. How few is not 
known, as HCFA does not collect data on complaints. Thirty-eight PROs that 
responded to a survey conducted by PRONET, an organization of PRO beneficiary 
representatives, reported receiving a total of 65 1 quality complaints between October 
1 . 1987, and the fall of 1988, with a range of to 97 complaints and an average of 
18 complaints.^^ Knowledgeable persons with whom I spoke, moreover, confirmed 



360 Id § XVIKAKS). ai 37 

361. Id. § XVII(AH7). al38. 

362. 42 C.F R. § 476 132(c) (1987). The incorporalion of 42 C.F R. § 476.132(c) is maiDUined by Ihe proposed 
regulations. 5^^ Protoseo Substandaud Care Rboulations. supra noie 61 . ai 1964 (lo be codiHed at 42 C F R. pi 462. 
§466 106(b)). 

363. Twiu) Score of Wcmx. supra note 3. § XVIl(A)(6)(a). at 37 

364 Id. § XVU(A)(6)(b). al 38 This will apparently be changed if the proposed regulations for review of 
beneflciary comptauits aie implemented, as they permit disclosure of physician-specific information to beneficiaries Srr 
PROrosEO SuBSTANDAM) Car£ Reculattons. supra note 61. at 1960. 1964 (to be codified at 42 C F R. p«. 476. subpt B. 
S476.l33(bH4)). 

365. PRONET. Beneficia«v WwrreN Quauty CoMPLAua Suhvev 2-3 (1988). 



MEDICARE PEER REVIEW 133 

the results of my own less formal survey: most PROs receive from five to forty 
complaints a month. 

The paucity of complaints is not surprising. First, few beneficiaries are aware 
that the PROs are available to receive complaints. Community outreach is a growing 
emphasis of the PROs. Under the Third Scope of Work, PROs must submit a detailed 
community outreach plan, which includes such comp>onents as a toll-free number 
(which several of the PROs already have), educational programs and seminars, and 
publication of informational materials.'** The PROs, however, still have far to go. 
Several representatives of beneficiary groups complained to me of the difficulties 
consumers have encountered in locating anyone at their PRO who is willing or able 
to take a complaint. One said that if a person from their organization calls a PRO to 
make a complaint, the PRO does not know what they are talking about. 

Second, the statutory requirement that the complaint be in writing undoubtedly 
deters many complaints. PROs responding to the PRONET survey reported receiving 
3036 total complaints by telephone, but only 944 in writing.'*' Medicare beneficia- 
nes with current or recent experiences of unsatisfactory medical care are often feeble 
and debilitated. Making a phone call to a PRO to present a coherent complaint 
requires a major effort. If the PRO tells the caller that the complaint can only be 
investigated if it is received in writing, the complainant must make a substantial 
additional effort, which probably will not be forthcoming. Many beneficiaries also 
interpret such a response to mean that the PRO is an uncaring bureaucracy, which is 
not interested in the complaint and will not take it seriously. 

One PRO with which I spoke attempts to deal with this problem by assisting the 
complainant in drafting the complaint while on the phone. HCFA or the PROs could 
also develop a form that could be immediately sent to complainants to fill out, with 
a stamped return envelope, or that could be completed by the PRO with information 
gained from the phone call.'** Only two PROs reported to PRONET that they assisted 
complainants by filling out a complaint form based on the verbal complaint and 
sending it to the complainant for verification and signature.'** A more sensible move 
would be to remove the requirement of a writing altogether. It serves no obvious 
function other than perhaps to deter some frivolous complaints. Though screening is 
undoubtedly necessary to eliminate vague or unsubstantiated complaints, PROs 
should be willing to investigate specific oral complaints. "^ 

Third, HCFA should amend its instructions to the PROs to assure protection of 
the confidentiality of complainants."' Under complaint investigation procedures 

366. Third Score of Work, supra nott 3. § XVI(C), at 34. 

367. PRONET. supra ncHe 365, al Al. This total includes both the quality-related and nonquality-related 
complaints. 

36K. See Lxller. supra note 172, at 2-3. One HCFA representative with whom I spoke afrirmed this idea but 
suggested the PROs do it as they, unlike HCFA. are not subject to the Paperwork Reduction Act and thus can develop 
new forms more quickly. 

369. PRONET. supra note 365. al Al. 

370. Cf. Cal. Health & Safety Code § 1419 (West 1979 & Supp. 1988) (complaints involving nursing homes may 
be submitted in writing, by telephone, or by persoiul visit: oral complaints shall be reduced to writing by the Department 
of Health): III. Ann. Stat. ch. Ill 1/2. 1 4153-702(a) (Smiih-Huid Supp. 1988). 

371. The Senate version of the complairu investigation provision included a requirement for protecting the 



134 TIMOTHY STOLZFUS JOST 

required by the Third Scope of Work, a provider or practitioner being investigated will 
be able to discover immediately the identity of the complamant because the PRO will 
request specifically the record of the complainant and a response to the complaint. 
But the PRO is required even to go beyond this and to request specifically the 
attending physician's opinion as to whether the complainant is sufficiently stable to 
receive information resulting from the complaint investigation. 

The exposure of the complainant's identity throughout the process not only 
serves to discourage complaints, but it may also actually put some complainants at 
risk for their health and safety. Two representatives of beneficiaries with whom I 
spoke noted that the PRO's practice of informing skilled nursing facilities (SNFs) of 
the identity of residents who had complained about conditions in the SNF could be 
dangerous to the complainant. The vulnerability of nursing home residents to physical 
and mental abuse is widely acknowledged^'^ and has been recognized in slate nursing 
home laws protecting the confidentiality of complainants.^'' Yet even when care in 
the hospital is involved, and abusive retaliation is less likely, the elderly, debilitated 
patient or her family may fear that care will suffer further if the provider or physician 
becomes aware that the patient has complained to the PRO. One interviewee noted 
the tenuous nature of contemporary physician/patient relationships and expressed the 
opinion that few beneficiaries would risk this relationship to complain of problems 
without strict protection of confidentiality. 

Given the broad pxjwers the PROs have to monitor care in institutions under their 
jurisdiction, they could easily investigate complaints while protecting the identity of 
complainants. A PRO, for example, could request a facility that was the subject of 
a complaint to provide the records of ten patients, including the complainant and nine 
others chosen at random. Indeed, a PRO could intensify review of a particular 
provider or physician, reviewing all cases or all cases in a particular area. If a 
problem were identified, the PRO could then approach the provider with a panern of 
cases, no one of which would be identifiably linked to a complainant. If it became 
absolutely necessary at some point in a proceeding to identify a complainant, the 
complainant could be offered the opportunity (afforded by a number of state nursing 
home laws) to withdraw the complaint rather than be put at risk.''* 

The Third Scope of Work provision that allows a physician target of a complaint 
to decide whether a complainant is capable of receiving information about the 



conridentiality of the complainant. This requirement was Jeleted without comment by the Conference Commi'tee. 5** 
CoNF. Rep. 1012. 99th Cong.. 2d Sess. 360-61. This does not necessanly picclude HCFA from establishing 
conHdeniiality lequirements. as they have elsewhere in the PRO program, see 42 C.F R. § 473, but may argue in favor 
of (he need for a statutory, and not just a regulatory, change. 

372. See Stannard. OU Folks and Dirty Work: The Social Conditions for Patient Abuse m a Nursing Home. 20 Soc. 
Prom. 329(1973). 

373. Cal. Health & Safety Code § 1419 (West 1979 & Supp. 1988); III. Ann. Stat. ch. Ill 1/2. 1 4153- 702(c) 
(Smith-Hurd 1988): Iowa Code Ann. § 135C.37 (West Supp. 1988); Mich. Comp. lJ^w$ Ann. § 333.2l799a<3) (West 
1980); ABA. Model Recommendations: Intermediate Sanctxjns for Ewporcement of QvAurr or Came in Nursing 
Homes 33. 37. 40(1981). 

374. See III. Ann. Stat. ch. Ill 1/2. 1 4153-702 (Sraith-Hurd 1988); Mich. Comp. Laws Ann. § 333.2l799a(3) 
(West 1980). 



MEDICARE PEER REVIEW 135 

investigation is particularly offensive, indeed bizarre, and it should be deleted as soon 
as possible. 

It is ironic that although the PRO Third Scope of Work offers no protection for 
the confidentiality of complainants, it goes so far in protecting the privacy of problem 
physicians that it actually violates the statute. This is a fourth problem with 
implementation of the beneficiary complaint investigation requirement. OBRA '86 
requires that the PRO "shall inform the individual (or representative) of ihe 
organization's final disposition of the complaint.""' The Third Scope of Work. 
however, ever solicitous to protect the reputation of physicians, only allows the PRO 
to give the beneficiary an evasive report that an investigation is being conducted and 
corrective action, if necessary, will be taken. This response is contrary to the 
requirements of the statute that the PRO explain its "final disposition.""* Moreover, 
it is likely to discourage complainants, who may be reluctant to pursue a complaint 
without any possibility of discovering its disposition. It is hoped that this problem 
will be cleared up by the proposed regulations, which seem to permit disclosure of 
physician-specific information.^^ 

Fifth, representatives of beneficiary organizations with whom I spoke also 
believe that beneficiaries are discouraged from complaining by the cumbersomeness 
of the process. For example, most PROs define "quality" issues quite narrowly and 
refuse a significant proportion of complaints lodged with them as inappropriate. ^^^ By 
contrast, two of the PROs with which I spoke pursued nearly all complaints lodged 
with them and were quite successful, for example, in straightening out billing and 
coverage disputes for their beneficiaries. This should be the norm, not the exception. 

Next is the problem of delay. Under the process outlined above, it can take the 
PRO nearly three months to investigate a complaint. Contrast this with state nursing 
home complaint investigation statutes requiring investigation of abuse and neglect 
complaints to be completed within seven days, or within twenty- four hours if a 
resident's life or safety is inmiinently threatened, and all other investigations within 
thirty days."* While the statute requires that practitioners and providers must be 
given "reasonable notice and opportunity for discussion" before the PRO concludes 
that their services were of poor quality, the length of time that is reasonable in any 
particular case should be determined considering the seriousness of the complaint.'*" 



375. OBRA 86. Pub. L. No. 99-509. 1986 U.S. CooeCowo. & Aomin. News ( 100 Sui.) 1874. 2047 (to be codified 
al 42 U.S.C. i I320c-3(a)). 

376. 42 U.S.C. S l320c-3(aK9) lequiics confidentialiiy of PRO dau generally and permiu ducloruie to the extent 
necessary "to cany out the puiposes of this pan." Here disclosure is clearly necessary. 

377. See supra note 364 Again, it is instnictive lo consider musing home statutes that not only require the 
investigative agency to give the complainant a full accounting of its investigation, but also allow the complainant an 
oppociunity to appeal if dissatisfied with the result. See III. Ann. Stat. ch. Ill 1/2. 1 4l53-702(g) (Smith-Hurd 1988); 
A.B.A. Model Rbcommendatx>ns. supra note 373. at 37. 

378. The PRONET survey discovered that some PROs consider nearly all complainu to be quality complaints, 
othen only a very small percentage. Two PROs classified only 3% of the complaints they received as quality complaints. 
PRONET. supra note 365. at 3. 

379. III. Ann. Stat. ch. Ill 1/2. 1 4153-702 (Smith-Hurd 1988). 

380. HCFA's recent proposed regulations require PROs to allow practiboners or institutions 30 days to respond to 
complaints prior (o notificaoon to beneficiaries, further slowing the process. See Protosed Substandaro Care 
RecuLAnoNS. supra note 61 . at 1964 (to be codified al 42 C.F.R. pt. 462. i 466. 106(a)). That PROs are capable of greaUy 



136 TIMOTHY STOLZFUS JOST 

Several beneficiary representatives complained more generally that the PROs 
will not communicate with complainants once the complaint is lodged and that they 
tend to dismiss complainants and complaints too readily. One stated that an individual 
beneficiary would seldom be able to pursue a complaint to a successful conclusion 
without strong support from a determined family or beneficiary organization. Others 
complained of the limited scope of PRO complaint investigations — they are generally 
restncted to the facility record and ignore care received in doctors' offices or 
following hospitalization. Finally, once PROs identify problems, there is often little 
they can do to solve them. With nursing homes or home health agencies, for example, 
the PRO may well end up referring a confirmed problem to the state licensing and 
certification agency, which could have handled the initial complaint more rapidly and 
with less danger to the complainant if the complaint had simply been forwarded to it 
in the first place. 

This final observation highlights a basic problem with the PRO complaint 
investigation procedure: recent federal and state consumer protection initiatives are 
building a reticulate, and sometimes redundant, system for beneficiary protection. 
OBRA '87, for example, requires investigation of complaints involving nursing 
homes and home health agencies at the state level. ^*' All states have physician 
licensure agencies, most license hospitals, all license nursing homes, all have nursing 
home ombudsmen programs. Until recently, the PRO role in this network of survey 
and enforcement agencies has been to monitor hospital data on an ongoing basis and 
to study and correct problems identified through that monitoring process. The 
complaint investigation requirements of OBRA '86 require PROs to take on an 
additional new identity, much more closely akin to traditional law enforcement. This 
may be appropriate in some settings, such as hospitals, in which PROs have extensive 
experience and in which there are few alternatives to which the beneficiaries can turn, 
it is less appropriate in the nursing home setting, in which other modes of state and 
federal regulation are more established and PROs have little experience. ^^^ PROs will 
not easily become comfortable in this new role, and if they are to take it on 
effectively, HCFA will have to give more leadership than it is currently offering.^"' 

VIII. Hospital Notices of Noncoverage 

With the DRG PPS Medicare reimbursement system came a fear that the new 
system would create incentives for hospitals to discharge patients "sicker and 
quicker "^*^ Because hospitals are paid en a per admission basis, it was thouglu they 



expediied review is demonsiraied by the notice of noncoverage procedures, discussed in Pan VIll below, under which 
decisions are rendcfcd in three days or less. 

381 OBRA '87. Pub. L. No. 100-203. §§ 4025. 4202(a)(2), 1987 U.S. Code Conc. & Admin. News (101 Stai.) 
1330, 1330-74. 1330-174 to -179 (to be codified ai 42 U.S. C. § 1395x(v)(l)(L), I395aa(d)). 

382. See iNSTiTUTt of MeoiaNE, supra note 304, at 146-70. 

383 A drali manual transmittal addrcssmg the PRO complaint investigation authority anempts to descnbe the kinds 
of problems PROs should address and those they should refer to others to investigate. This is a step in the nght direction. 
In fact. onl> j9 of the 517 classiHed quality complaints received by the PROs identified in the PRONET survey pertain 
to nursing homes. PRONET. supra note 365, at Al. 

384. See E-xamirtalwrt of Quality of Care Under Medicare's Prospective Payment S\slem: Hearings Before the 



MEDICARE PEER REVIEW 137 

would attempt to discharge patients as quickly as feasible to reduce their costs and 
increase their profits. The PROs, as Medicare's primary bulwark against erosion of 
the quality of medical care, were quickly brought in to assist in assuring that 
Medicare beneficiaries were not discharged until discharge was medically appropri- 
ate. Resulting legislation, regulations, and manual provisions have created a tembly 
complex process that has left beneficiaries confused and their representatives 
frustrated. 

OBRA '86 created the basic framework under which PROs review hospital 
notices of noncoverage. If a hospital determines that a Medicare beneficiary no longer 
needs inpatient care, and the patient's attending physician agrees, the hospital may 
give the patient a notice of this determination.^" This notice is called a notice of 
noncoverage, because it is effectively a notice that hospitalization is no longer 
necessary and, therefore, no longer covered by Medicare.^** If a hospital gives this 
notice, and the patient refuses to leave, the hospital may begin charging the patient 
directly for his or her care after the second day following the date of the notice (after 
two "grace days").^*' 

The patient may, however, request the PRO to review the hospital's determi- 
nation. If the patient requests the review no later than noon of the first working day 
after the date he or she receives the hospital notice, the hospital must provide the PRO 
with the patient's records by the close of that day.'** The PRO must then review these 
records and discuss the discharge with the patient and the patient's attending 
physician.^*' The PRO must complete its review by the close of the next full working 
day after it has received the records^^ and then send a notice to the patient and 
hospital of the results of its review. If the patient makes a timely request for review, 
the hospital may not charge the patient for inpatient hospital services received prior 
to noon of the day after the patient receives notice of the PRO decision. 3^' Of course, 
if the PRO decides that discharge is medically inappropriate, the hospital cannot 
charge the patient until discharge is appropriate.^^ 

If the hospital determines that inpatient care is no longer needed, but the 
attending physician disagrees, the hospital may ask the PRO to review its 
determination. ^^^ The hospital may not charge the patient until the PRO rules in its 
favor. If the PRO rules in favor of the hospital, and against the attending physician, 
the hospital may issue a notice of noncoverage. The patient may then request the PRO 
to reconsider its earlier decision. The PRO must complete this reconsideration within 



Senate Comm. on Fin. . 99th Cong. . 2d Sess. 1 14. 142. 276. 2:)0 (1986); Quality of Care Under Medicare s Prospniive 
Payment System: Hearings Before the Senate Special Comm. on Aging. 99th Cong.. Isl Sc$$. 363-65 (19861 (Vol. I). 

385. 42 use. § 1320c-3(e)(l) (1982 & Supp. IV 1986). 

386. Id. 

387. 42C.F.R. § 412. 42(cK3)(ii) (1987). 

388. 42 U.S.C. § 1320c-3(e)(3)(B) (1982 & Supp. IV 1986). 

389. Id. § 1320c-3(eK5); PRO Manual, supra note 53. §§ 1M4I20. IM4130. 

390. 42 use. § 1320c-3(e)(3KB) (1982 & Supp. IV 1986). 

391. Id. 5 l320c-3(eK4). 

392. 42C.fr. § 412.42(c)(4) (1987). 

393. 42U.S C § 1320c-3(e)(2)( 1982 & Supp. IV 1986). Under a provUion of OBRA 87, it must notify the patient 
thai it has done so. 



138 TIMOTHY STOLZFUS JOST 

three days.'** As the hospital may begin charging the patient after the two grace days 
elapse, however, the patient may have to pay for the fmal day of the reconsideration. 

The Medicare statute also requires that hospitals give beneficiaries a description 
of these appeal nghts at the time they are admitted. '*" The Medicare Hospital Manual 
requires that this statement be a copy of a document entitled "An Important Message 
from Medicare." The PROs are to monitor hospitals to be sure that they, in fact, give 
beneficiaries this notice.*'* Under a rule proposed in June 1988. the hospital would 
have to "obtain a separate signed acknowledgement from the beneficiary attesting to 
the receipt of the statement, and maintain a copy of the acknowledgement."*''^ 
Finally, the PRO is responsible for reviewing all cases in which the hospital charges 
a patient for care after a notice of noncoverage is sent.*'* If the PRO determines that 
the notice of noncoverage was inappropriate, the hospital must refund inappropnately 
collected charges.*'' The PRO regulations and Medicare Hospital Manual also 
provide for an expedited three-day appeal process when a hospital or PRO denies a 
patient admission to a hospital on the grounds that the admission is not covered by 
Medicare because it is not for necessary care. ■**'*' 

If the procedures described in the preceding paragraphs seem confusing to the 
reader, imagine how they must seem to an elderly Medicare beneficiary who has just 
been admitted to a hospital or is just about to be prematurely discharged. Beneficiary 
representatives witn whom I spoke uniformly complained that HCFA's "An 
Important Message from Medicare" was unintelligible to many beneficiaries and, 
until recently, in error. Moreover, most beneficiaries are unaware of the "Important 
Message," which is generally handed to them with a sheaf of other admission papers 
at a time when they are obviously distracted by other concerns. One recent survey 
revealed that 82% of beneficiaries claimed that they had not received the notice at 
admission.^' An AARP study of persons over sixty-five who had been hospitalized 
in 1987 found that more than half of the respondents (56%) either believed that they 
had not received the notice or did not know whether they had received it or not.'*^ 
Several beneficiary representatives with whom I spoke claimed that hospitals seldom 
give patients notice of noncoverage, preferring to work with the doctor to convince 
the patient to leave without threatening to make the beneficiary pay for the care.-"** 



394 42 C.F R § 473.32(a)(1) (1987). 

395 42 U.S.C. % l39Sc<:<aKIKM) (1982 St Supp. IV 1986). These provisions are implemenied by 42 C.F R. 
§ 466 78(bM3). 

3%. y.EDKKK. Ilosp. M\NVAL, supia note 56. § 312.1; PRO Manual, supra note 53. § 2(X)S 

397 53 Fed. Reg. 22.523 (1988) (to be codified at 42 C.F.R. § 498.27, (proposed June 16. 1988). 

398 PRO Manual, supra note 53. § IM2080. 

399 42 C F R. § 412. 42(cK3Kv) (1987). 

400 See id. § 473 32(aM I); Wilson. How to Appeal Medicare Hospital Coverage Denials Under the DRG System. 
20 Cleajilnghouse Rev 434 (Sununer 1986). 

401 Conversation with Laura Schoenberg. Pennsylvania Public Intetcst Coalition. 

402 AARP. OnAfT OF Survey by Markettacts 5 (1988) 

403 In fact. PRO data reveal that the extent of issuance of hospital notices vanes sigmricanily from PRO lo PRO. 
New York hospitals had issued 39.726 notices as of June 1988. Washington hospitals only 10 The extent of PRO 
disagreement with the hospital also varied signincantly. from a disagreemenl rate of 41% in Michigan and 44% in 
I>laware.io3%inMinnesou. Montana, and Maine. Health Care Fin. Admin.. U.S. Dep't of Health & Human Servs.. 
MotmtLY Peek Review Omsanizatxjn (PRO) Data Summary D3 (Sept. 8. 1988) |hereinafter PRO Data Summary). 



MEDICARE PEER REVIEW 139 



Though hospitals must all disseminate the "Important Message" at admission, they 
can draft their own notices of noncoverage to give at discharge, and these vary 
significantly, some not explaining the patient's rights at all. 

The most important issues in this area concern whether the beneficiary should 
always be entitled to a liability-free appeal to the PRO, when the beneficiary should 
be given a notice of his or her rights, and what the notice should say. As described 
above, if the hospital notifies the beneficiary that it and the beneficiary's physician 
concur that the beneficiary should be discharged, and the beneficiary appeals by noon 
of the next day, the beneficiary cannot be charged for care until the PRO rules on the 
appropriateness of the discharge. If, however, the hospital and the physician 
disagree, the hospital requests PRO review, the PRO rules in favor of the hospital, 
the hospital issues a notice of noncoverage, and the beneficiary requests PRO 
reconsideration, the beneficiary may end up paying for a day of care while awaiting 
the PRO decision. 

This two-track process is responsible for much of the complexity of the 
"Important Message" and the confusion of the current process. Surely the cost to the 
hospitals of the extra day of care in the latter situation, which cannot be too common, 
does not justify this added confusion and complexity. Alternatively, PROs reviewing 
beneficiary appeals, when they have already ruled on the disagreement between the 
doctor and hospital, should be able to reconsider their decision within the two grace 
days, because they already have the records and have talked to the parties during the 
prior review. In any event, it should be possible to provide the beneficiary with a 
liability-free review by the PRO. OBRA '86 only provided a liability-free appeal 
from notices of noncoverage when the hospital and physician concurred. OBRA '87 
appeared to assure a liability-free appeal for all beneficiaries, regardless of physician 
concurrence or nonconcurrence in the original decision, but a technical correction to 
OBRA '87, found in the Catastrophic Coverage Act, restored the two-track 
system.**** The statute should be amended to assure that every beneficiary who 
receives a notice of noncoverage, and who appeals by noon of the next day to the 
PRO, is entitled to remain in the hospital without additional charge until the PRO 
decides the appeal. 

If it is the intent of HHS to make beneficiaries aware of their rights, it is 
troubling to rely on information concerning these rights given beneficiaries at the 
time of admission, when most beneficiaries (and their representatives) are over- 
whelmed with the anxiety of a hospital admission and other paperwork connected 
with the hospital stay. Even having the beneficiary or representative sign the notice 
will help little, as the notice will remain only one more paper to be signed as part of 
the admission process. Notice of beneficiary rights should, therefore, also be given 
as part of the discharge planning process. ***' 



404 Medicare Catastrophic Coverage Act of 1988. Pub L. No. 100-360. § 4096. 1988 US. Code Cong. & Admin 
News (102 Slat.) 683, 791 (to be codified at 42 U.S.C. § 1320c-3(eM3HAMi). 1320e-3(eM2HB)) (to be amended by a 
subsequent technical conection striking "before 'paragraph (2)' " in Family Support Act. Pub. L. No. 100-485. § 608, 
1988 U.S. Code Cong. & Admin. News (102 Stat.) 2343, 2421 (to be codiHed at 42 U.S.C. § 1320c-3)). 

405. Section 9305(c) of OBRA 86. Pub. L. No. 99-509. 1986U.S. Code Cong. & Admin. News (100 Stat.) 1874. 



140 TIMOTHY STOLZFUS JOST 

At the time a hospital or attending physician determines that a Medicare patient 
no longer needs hospital care, the patient ought to be given notice of this decision and 
of his or her rights under the law. This information should be given the patient as soon 
as possible after it is determined that the patient is ready for discharge and at an early 
enough point to allow the patient to exercise his or her rights. The notice should be 
provided to the beneficiary by his or her own physician or by an employee of the 
hospital responsible for discharge planning. Alternatively, the notice could be posted 
on the wall near the bed in all patient rooms that are used for Medicare beneficiaries, 
so that the beneficiary could read it at his or her leisure. While providers may object 
that this procedure may lead to more appeals, it is hard to sympathize with an 
argument that beneficiaries should be kept ignorant of their rights to keep them from 
burdening the system through the exercise of those rights. 

The notice given the patient should be as simple as possible.'"'* It need only tell 
the patient that, even though the hospital has decided that discharge is appropriate, 
the patient can stay and not be charged for the stay until after the second day 
following the date of the notice. If the recommendation to merge the two-track system 
into a single system is adopted, the notice would further state that, should the patient 
decide to appeal and promptly calls (by noon of the next day) the PRO (whose 
toll-free number would appear prominently in the notice), the patient could not be 
discharged or charged for care until the day following the PRO's decision on the case. 
If a two-track system is retained, the notice just described could be given when the 
doctor and hospital agree, and a separate, differently worded notice could be given 
if the hospital has already received review from the PRO. This notice would state that 
a prompt appeal to the PRO would secure review within three days, with the patient 
liable for only, at most, one day of care. This is all the beneficiary really needs to 
know, and further information may be more confusing than helpful. This notice (or 
notices) should be on forms provided the hospital by HCFA to assure uniformity and 
accuracy. It should be drafted with ample input from beneficiary organizations to 
assure that it is comprehensible to beneficiaries. 

IX. Utilization Review Deniai^, Reconsiderations, and Appeals 

While PROs do not yet have regulatory authority to deny payment for 
substandard care, they regularly exercise their authority to deny payment for care they 
determine to have been rendered unnecessarily, or in an inappropriate setting, or to 
partially deny requested payment by modifying the DRG assigned by a hospital.**" 
When PROs determine that payment should be denied on the basis of lack of 
necessity, they are required further to determine whether beneficiaries, practitioners, 
or providers should be excused from liability for the cost of this care because they did 



1989, requires hospitals to assist Medicare patients with discharge planning. 42 U.S.C. § l39Sx(ee) (1982 & Supp. IV 
1986) (implemented by regulations proposed at S3 Fed. Reg. 22.506 (1988)) 

406. Ser Lener. supra note 172, at 4. 

407. 42 U.S.C. § I320c-3(a) (1982 & Supp. IV 1986); 42 C.F R. § 412 60(d) (1987): id. pi. 466 



MEDICARE PEER REVIEW 141 



not know, and could not reasonably have known, that Medicare would not pay for the 
care.^* 

The PRO utilization review denial process is normally initiated when" a nurse 
reviewer, applying screening criteria, kicks out an aberrant medical record. This 
record is then reviewed by a physician (usually not of (he specialty of the physician 
whose work is being reviewed) who then sends a notice to the attending physician and 
to the provider requesting an explanation. -"^^ The PRO must allow the physician and 
provider twenty days to respond to the query and must also allow an opportunity to 
discuss the problem by telephone.*'" The provider and physician response is then 
reviewed again, and a final initial determination is made by the PRO. In several of 
the PROs I interviewed, this determination is made by the initial reviewer, in others 
it is performed by a second reviewer. In only one of the PROs I interviewed was there 
an attempt to use a matched specialist for this review.*" Notice of this initial 
determination is sent to the beneficiary, physician, provider, and the fiscal interme- 
diary (which denies payment to the provider). *'2 

A beneficiary, provider, or physician affected by an initial denial, unfavorable 
waiver of liability determination, or DRG change may request a reconsideration by 
the PRO within sixty days of receipt of notice of the denial.*'' The case is then 
reconsidered by a matched board-certified or board-eligible specialist, who cannot be 
the person who made the initial determination.*'* If a beneficiary requests a 
reconsideration, the PRO must complete its review within three working days if the 
beneficiary is an inpatient in a hospital (and the claim relates to the hospitalization) 
or if the initial determination was a preadmission review and denied admission, or if 
the patient is an inpatient in a skilled nursing facility, the PRO must complete its 
review within ten working days.*" All other PRO reviews must be completed within 
thirty working days.*'* Some PROs offer an evidentiary hearing at the reconsidera- 
tion level; others conduct a review on the written record (considering any additional 
information offered by the requester) or by telephone.*'^ 

If the PRO reconsideration decision is unfavorable, and the amount in 
controversy involves more than $200, the beneficiary may appeal the decision to an 
ALJ.*'* If more than $2000 is involved, the beneficiary may seek Judicial review. *■' 

408. 42 use. § I395pp (1982 &. Supp. IV 1986); 42 C.F.R. § 466.86(»K4) (1987). 
409 42C.F R. § 466.93 (1987). 

410. Thiiid Scok of Womc. supra note 3. § lll(B). at 2-3. 

411. The regulaiiofu only require thai the detenniiutton be made by a physician. 42 C.F.R. § 466.98. though the 
PRO Marual olu requires tiiai it be done oy a physician who praaices in a seitirg similar to that of the physician whose 
care is being reviewed. PRO Manual, supra note 53. 5 IM2(X)2. The AMA PRO Executive Survey, supra note 136. 
Table 7. found that 28% of the PROs use specialists for initial medical reviews. More infotmation on this is contained 
in a recent GAG study. See generally GAG Medkake PROs. supra note 17S. 

412. 42 C.F.R. § 466.94(a) (1987). 

413. 42 use. § I320C-4 (1982); 42 C.F R. §8 473. 15. . 16. .20 (1987); PRO Manual, supra note 53, § 4082. See 
generally Wilson, supra note 400 (on PRO appeals). 

414. 42 C.F R. § 473.28 (1987); PRO Manual, supra note 53. § 4084(A). 

415. 42 C.F.R. 5 473.32(1987). 

416. Id. 

417. PRO Manual, supra note 53. $ 4086(B). 

418. 42 use. § I320c-4(I982). 

419. ;</. 



142 TIMOTHY STOLZFUS JOST 

A provider or physician denied payment, however, has no recourse beyond the 
reconsideration, except the right to appeal unfavorable waiver of liability 
determinations. *2° The provider or practitioner may also request an ALJ hearing if the 
PRO found it to have inappropriately discharged or transferred a patient in 
circumvention of the PPS system."^' The physician or provider cannot collect from 
the patient if payment is denied. 

As a practical matter. PRO determinations seldom involve large enough sums of 
money to justify the cost of an appeal to an ALJ or judicial review. '•^z Moreover, 
smce only the beneficiary can directly appeal a PRO necessity or appropriateness 
decision, and since the beneficiary is usually excused from payment for denied care 
under the waiver of liability provisions, there is seldom sufficient incentive for the 
beneficiary to appeal. 

PRO officials with whom I spoke generally felt the utilization review system was 
working rather well. Beneficiary and provider representatives were less sanguine, 
though few put revising the utilization review process at the top of their PRO agenda. 
Several concerns, however, were raised by these representatives. 

Fu^t, providers and practitioners are concerned that they have no means of 
directly appealing an adverse PRO utilization review decision to an ALJ. 42 U.S.C. 
section 1 320c -4 only allows beneficiaries the right to appeal PRO decisions to an 
ALJ. As noted above, providers may directly appeal adverse waiver of liability 
decisions, but not the underlying decision on necessity or appropriateness (except 
when the provider has been found to have taken an action to circumvent PPS).-*^^ 
Even if a provider wins a waiver of liability appeal and the underlying decision is 
wrong, the provider will be bound by that decision in future cases, for it will not again 
be able to argue that it did not have notice of the lack of necessity of the service, a 
necessary element of a waiver of liability determination. *2* Section 93 1 3(a) of OBRA 
'86 permitted providers and practitioners to represent beneficiaries in Medicare 
appeals (if they did so without charge and waived any right to payment for the 
services that were the subject of the appeal), "^s thus effectively allowing providers 
and practitioners direct appeal rights when the beneficiary agreed to allow such 
representation. This provision, however, expressly does not apply to PRO 
decisions. ••2* The Senate version of OBRA '87 extended it to the PRO program, but 



420. Id. i I395pp«l) ( 1982 &. Supp IV 1986) This appeal is only available if the benericlary docs not appeal and 
if the amount in controversy exceeds $100 for Pan A determinations or $500 for Pan B determinations. Judicial review 
IS available for waiver of liability determinations involving more than $1000. 

421. PRO Manial. supra note S3. § 4088(A): S3 Fed. Reg 8.665 (1988) (to be codiHed at 42 C.F.R. § 466.83) 
(proposed Mar. 16. 1988). This distinction is often no< clear. A PRO could find an inapprx>pnate readmission. for 
example, to be not medically necessary or to be the result of a premature discharge. 

422. Some of course will, .'iuch as an admission denial involving a costly medical procedure. 

423 PRO Ma-sxal. supra note 53. $ 4088(A). 

424 42 U S.C. § l395pp<aM2)(l982). 

425 OBRA '86. Pub. L. No 99-509. I986U.S. CooeConc & Aomin. NewsdOOSut.) 1874. 2002 (to be codified 
at 42 use. 5 l395ff(bKl)). 

426. 42 U.S.C. § l395ff(bKIKD) (1982 & Supp, IV 1986) 



MEDICARE PEER REVIEW 143 

this amendment was dropped from the conference bill/^' and thus providers and 
practitioners remain without any means of appealing PRO decisions. 

Because few PRO utilization review denials involve large enough sums to justify 
the cost of an appeal (or even to reach jurisdictional amounts for an appeal), providers 
and practitioners are much more concerned that utilization review decisions be made 
correctly in the First place. The data are not encouraging. Nationally as of June 1988 
during the second contract cycle, 3.862% of reviewed cases were denied. '♦^s In over 
30% of these cases, a reconsideration was requested. '♦^^ In nearly 44% of these cases, 
the initial decision was reversed on reconsideration. Reversal rates of individual 
PROs ranged from 10% to 71%.'»«> The SuperPRO has also disagreed with PRO 
utilization review decisions in a significant proportion of cases. '•^i It is difficult to 
avoid the impression that PROs are often sending out initial denials without adequate 
consideration, trusting that if the decision is wrong it can always be changed at the 
reconsideration level. This is not a defensible approach under any circumstance. 
Considering that denial notices go to beneficiaries as well as to providers and 
practitioners, and often leave beneficiaries confused, anxious, and uncertain of their 
own liability for the denied care, this practice is inexcusable. 

Two approaches to improving the quality of utilization review decisions are 
apparent. First, more contact can be required between the PROs and the providers and 
practitioners whose records they are reviewing. This approach is taken by OBRA '87, 
which affords a practitioner or provider an opportunity of discussion and review of a 
proposed denial before the decision is implemented. '♦'^ Although this approach is 
sensible and is already the practice of many PROs, going further in the direction of 
requiring the reviewer to justify his or her decision to the practitioner or provider is 
not without costs. One PRO executive, for example, noted that it is difficult enough 
to get qualified reviewers, and it will be even more difficult if reviewers are expected 
to argue face-to- face with other doctors over every review decision. 

Second, specialist review can be required at the initial determination level. 
Currently, final initial decisions are made in most PROs by doctors who are not 
specialists in the type of medical care under review. '♦'^ Only at the reconsideration 
level is specialty-matched review required.*^ Given the highly specialized nature of 
modem medicine, this practice is unacceptable and goes far towards explaining the 
high level of reconsideration reversals. ^'^ It may not be practical to assign initial 



427. H.R. CoNF. REf. No. 495, lOOih Cong.. Isi Sesi. 666. rep-imet' in 1988 U.S. Code Cong. & Admj<. News 
2J13-1245. 23l3 1412. 

428. PRO Data Summary, supra note 403. Table Dl I. 

429. Id. 

430. Id. See Whalen, Schmitt. & Rossetti. Early Experience with Peer Review Organizations. 3 J. Gen. Internal 
Med. 59 (Jan. -Feb. 1988) (describing the PRO reconsideration process from a reviewer's perspective). 

431. GIG Ejtectiveness Retort, supra note 9. at 12. 61. 

432. OBRA 87. Pub. L. No. 100-203. § 4093. 1987 U.S. Code Cong. & At>MiN. News (101 Stat.) 1330. 1330-136. 
See 53 Fed. Reg. 8.865 (1988) (to be codified at 42 C.F.R. § 466.93) (proposed Mar. 16. 1988). 

433. See 42 C F R. § 466.98 (1987). 

434 See id. § 473.28: 53 Fed. Reg. 8.666 (1988) (to be codified at 42 C.FR § 473 28) (proposed Mar. 16. 1988): 
PRO Manual, supra note 53. § 4084. 

435. See Whaien. Schmin. A Rossetti. supra note 430. at 61. 



144 TIMOTHY STOLZFUS JOST 

reviews and the drafting of proposed decisions to matched specialists. Indeed, if the 
review is done on-site, it is probably not possible. But the final initial decision, made 
after the provider or practitioner has had an opportunity to review and respond to the 
proposed decision, should be made by a specialist. 

A final issue that is beyond the scope of this Article should be noted because it 
was mentioned so often by the provider representatives that I interviewed. The direct 
consequence of a PRO utilization review denial is that the fiscal intermediary is 
notified, which then denies reimbursement to the hospital under Part A for the 
unnecessary or inappropriate care. The intermediary should notify the carrier, which 
should deny payment to the physician who ordered the inappropriate care.'*^* 
Hospitals firmly believe, however, that this A-B link has not been made, and earners 
are not denying payment to doctors. Hospitals are exceedingly annoyed that they are 
routinely denied payment for unnecessary care, but the doctors who ordered the care 
are not p>enalized by Medicare in any way. Their annoyance seems justified, and 
taxpayers, who are financing the unnecessary care, should be equally upset. 

X. PRO Data Confidentiauty and Disclosure 

One of the primary functions of the PROs is to assemble data. Not surprisingly, 
therefore, they possess vast quantities of data on Medicare providers, practitioners, 
and beneficiaries. PROs are not subject to the requirements of the Freedom of 
Information Act.*" Rather, disclosure and maintenance of confidentiality of PRO 
data are governed by the PRO statute'*'* and regulations. **'' The general effect of 
these provisions is to protect data that identifies patients, providers, and practitioners 
against disclosure, except insofar as this information is required by federal or state 
regulatory and law enforcement agencies to carry out their responsibilities. 

The PRO confidentiality and disclosure provisions are generally consistent with 
recommendations made by the Institute of Medicine after a thorough study in 
I9g] 440 This study thoughtfully weighed the considerations in favor of disclosure 
and of confidentiality and reached conclusions that continue to be on the whole 
acceptable. I would favor greater disclosure of health care quality data to assist in 
creating a smoother functioning market for health care services,**' but this preference 
is grounded on health care policy rather than administrative law considerations.**^ 

Several confidentiality and disclosure issues have been discussed earlier in this 



436. Pan B Carrier's ManiuJ. i 4169: Goxrield. Ulilizaiioi, Managemeiu and Quality Assurance, 2 Meo. Staff 
CouNS. IS. 16 (Spnng 1988). 

437. 42 use. § I320c-9(a) (1982 & Supp. IV 1986). 

438. Id. § 1320C-9. 

439. 42 C.F R. pt. 476 ( 1987); ue also S3 Fed. Reg. 8.6S4. 8.666-67 ( 1988) (to be codified at 42 C.F.R. p(. 476) 
(proposed Mar. 16. 1988). 

440. See Instttvte of Medicine. Access to Medical Review Data: Disclosure Policy fou Pi«o«=essional Standards 
Review Okcanoations (1981). summarized in Searches for A Balance m Medical Data Disclosure: Report of a Study. 
I Health Aff. 62. 62 (Spnng 1982). 

44 1 . See Josl. The Necessary and Proper Role of Regulation to Assure the Quality of Health Care. 23 Hows. L. Rev. 
52S. 594-97(1988). 

442. Greater disctosure is favored also by beneflciary groups. See Lxner. supra note 172. ai 5-6: Letter from Sidney 
Wdfe to HCFA (July S. 1984) (conunenting on proposed PRO confidemiAtiiy and disclosure regulations). 



MEDICARE PEER REVIEW 145 

Article.**' There are two additional areas, however, in which PRO policy should be 
clarified to assist in effectuating the PRO mission. First, PRO sharing of data with 
state medical boards should be improved. The OIG study on PRO quality assurance 
activities found that 67% of the PRO CEOs interviewed believed that the relationship 
between their PRO and state medical board was relatively poor, and 27% felt that 
greater clarification was needed as to federal restrictions on sharing information with 
the medical board.*** This is consistent with my own discussions with the twelve 
PROs 1 interviewed, most of whom exchanged little data with the state medical 
boards. 

Section 9353(d) of OBRA '86 requires PROs to share data with state regulatory 
authorities. Regulations implementing this provision have been proposed by HHS.*** 
HHS should promptly promulgate regulations, issue Manual provisions implement- 
ing these regulations, and amend its Scope of Work to require the PROs to formulate 
a plan for sharing data with state medical boards.**^ 

The second issue involves PRO notification of hospitals of problems involving 
physicians on the staff of those hospitals. The PRO confidentiality regulations require 
PROs to disclose to an institution, on request, information regarding the practice and 
performance of practitioners in that institution.**' Some PROs with which I spoke 
have interpreted this provision liberally, notifying hospitals immediately whenever 
they identify quality problems involving physicians on their staffs and fully involving 
the hospital in the corrective action plan.*** Others have refused to release any 
information to hospitals about a practitioner without the consent of that practitioner, 
which is seldom given. The California Medical Association, recognizing the valuable 
role that hospital medical staffs can play in assisting physicians with practice 
problems, has urged HCFA to clarify the responsibilities of PROs to involve hospital 
medical staffs in resolving physician quality problems.**^ The AMA has debated 
taking a position favoring disclosure of physician problems to hospital medical staffs, 
but it stopped short of doing so.*^ The AHA has also asked that hospitals be notified 



443. See supra text at note* 278-81. 371-77. 

444. OIG Quality Refokt. supra note 167. at 18: see also QIC Licensure & Disopune Report, supra note 337, 
ai IS, 17 (faulting the state medical boards for their poor relationship with the PROs). See also GAG Medicare PROs, 
supra note 1 75, at 30 (S7% of PROs reported no contact with state medical boards in the fourth year of the program; only 
27% reported more than one contact). 

445. 53 Fed. Reg. 8.654. 8,666-67 (1988) (to be codified at 42 C.F.R. pt. 476) (proposed Mar. 16, 1988) HCFA 
should also alter its proposed regulaiioiu to conform to the law. 42 U.S.C. } I320c-9(b)(2). as amended by OBRA '86. 
requires disclosure on request to stale I'crnsure bodies of data or infonnalion "relating to a speciHc case or lo a pos<iible 
pattern of substandard care." The proposed regulatio>i requires disclosure on request of infoimation "relating to a specific 
case £>/a possible pattern of substandard care." S3 Fed. Reg. 8.654. 8.667 (to be codified at 42 C.F.R. §476. 138(a)(1)) 
(proposed Mar. 16, 1988) (emphasis added). As the tule would seem to require a pattern of substandard care before 
information regarding a specific case could be disclosed, it would permit sharing of much less information than the statute. 
I am infotmed that this is a misprint, which will be corrected by HHS in the final rule. 

446. See OIG Qualtty Report, supra note 167, at lii; current requirements in the Scope of Work require this only 
obliquely {see Third Scope of Work, supra note 3. §$ XVIII(G). XX. at 41-42). 

447. 42 C.F.R. § 476. l33(aK2) (1987). 

448. See AMA PRO Executive Survey, supra note 136. Table IS (85% of the PROs report meeting with medical 
staff to share corrective action plans). 

449. See Letter from Uurens White. CMA, to Thomas Morford. HSQB. at 3 (Mar. 30, 1988). 

450. PROs Draw Fire from AMA House, Med. World News. Jan. 12. 1987. at 26. 



146 TIMOTHY STOLZFUS JOST 

whenever a quality problem is identified by the PRO involving a physician on their 
medical staffs.**' 

Several HCFA memos issued in 1988 have further muddied the waters as to the 
PROs' responsibilities in this respect. A Regional Medical Review Letter issued in 
January 1988 informed the PROs that absent permission from a physician they could 
not disclose information concerning quality problems to a hospital medical staff 
unless the PRO identified a "pattern" of poor practice (involving more than one 
confirmed problem) and then only if the hospital requested the information. A 
subsequent Regional Medical Review Letter, dated June 1988, informed PROs that 
HCFA had reconsidered and that they could inform hospitals of any confirmed quality 
problems involving physicians on that hospital's staff, with or without a request. This 
reconsideration was based on the realization that any quality problem involving a 
physician potentially involved the provider as well. The June 1988 letter, however, 
prohibited PROs from divulging to hospitals corrective action plans they were 
considering. This left the PROs in a quandary, as most physician corrective action 
plans involve hospital oversight. A memo from HSQB, dated July 22, 1988, 
addresses this situation by permitting PROs to notify hospitals of corrective action 
plans when the hospital's participation is necessary to implement the plan or if the 
plan will affect the hospital. Otherwise, notification to the hospital is still 
prohibited. *'2 

HHS should amend its confidentiality regulations to allow PROs to disclose to 
hospital medical staffs information regarding any confirmed quality problems 
involving practitioners on the staffs of those hospitals and any corrective action plans 
involving those practitioners. Hospitals already have a pretty good idea of who on 
their staff is in trouble with the PROs because they know whose records the PROs are 
requesting. Explicit sharing of PRO information with the hospitals would help them 
to better focus their efforts in assisting these problem doctors to improve their 
practices. Such sharing is certainly consistent with the PRO statute, which permits 
disclosure of confidential information "to the extent that may be necessary to carry 
out the purposes of this part" and when the Secretary permits disclosure by 
regulation.*'^ This change would assist PROs in carrying out their quality assurance 
role, assist hospitals in dealing with problem doctors, and, ultimately, assist doctors 
in addressing quality problems before beneficiaries are harmed and sanctions become 
necessary. 



45 1 . See Letter from Jack Owen to William Roper (May 16, 1988) (commenting on proposed PRO regulations). See 
also Remarks of Buetel. Dedic. and Weil on the role of a hospital PRO committee in assisting physicians. Tianscnpc, 
Confeience on Understanding the Federal P.R.O.. 95-107 (Feb. 28. 1987) (sponsored by LoyoU University Health Law 
Institute. (Zhicago Academy of Medicine, and Resunection Hospital. Chicago). 

452. Memo from Richard Husk. Oirector OfTice of Peer Review, to John L. Setman. Associate Regional 
Administrator. HSQB. Regioa VU. 

453. 42 U.S.C. i 1320c-9(aKIH2) (1982 & Supp. IV 1986). 



MEDICARE PEER REVIEW 147 

XI. Conclusion 

The PRO program is still very young — a person its age would still be in 
kindergarten. During its short life it has undergone frequent and dramatic change. Not 
surprisingly, the beneficiaries , providers, and practitioners affected by the program 
have frequently found it frustrating and almost always confusing. Among those 1 
interviewed, however, there seemed to be an emerging consensus that the program is 
worth keeping and that it is maturing. The recommendations delineated in this Article 
are offered to assist in that maturing process, to fme-tune rather than to radically 
change the program. They will undoubtedly not be accepted by consensus. The 
parties affected by the program too often have already staked out positions too sharply 
at odds to make consensus possible. These suggestions are offered, however, with 
good will towards the program and to all those affected by it. It is hoped that those 
with the power to implement these recommendations will consider them carefully as 
a means towards improving the PRO program to better serve the Medicare program 
and its beneficiaries. 



REPORT TO THE 

ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

RECOMMENDING THAT 



GOVERNMENT CONTRACTING 
OFFICERS SHOULD MAKE GREATER USE OF 
ADR TECHNIQUES IN RESOLVING CONTRACT DISPUTES 



BY 

RICHARD J. BEDNAR, ESQUIRE || 

CROWELL & MORING 
WASHINGTON, D.C. 

I 
MARCH 1989 i| 



ACKNOWLEDGEMENTS 



This report was prepared for the consideration of the Admin- 
istrative Conference. The views expressed are those of the 
author, and so not necessarily reflect those of the confer- 
ence or its Committees. 



GOVERNMENT CONTRACTING 

OFFICERS SHOULD MAKE GREATER USE OF 

ADR TECHNIQUES IN RESOLVING CONTRACT DISPUTES 

INTRODUCTION 

To the reader whose experience in the federal Government con- 
tracts disputes process is extensive, the proposition embodied in 
the title of this report is hardly a shocker.^ Those familiar 
with the disputes process authorized by the Contract Disputes Act 
of 1978^ (CDA), know that its essence is not so much to provide 
direction and control from above, but rather to provide a 
framework within which the contracting parties themselves have the 
dominant role of determining their rights and duties within a 
particular contract. The gist of the CDA is that contract 
disputes first are presented to the Government contracting officer 
(CO) for resolution and aire litigated only if that initial effort 
fails . The goal of the CDA is to provide an informal and 
expeditious 3 process for resolving disputes without disrupting 
contract performance. Practitioners in this specialty generally 
agree that more cases should be disposed by agreement between the 
parties by whatever honorable, fair and legitimate means. ^ For 



This report addresses the process of resolving disputes which 
arise during performance of Government contracts and does not 
extend specifically to controversies which arise during the 
contract formation process . 

41 U.S.C. SS 601-613; 5 U.S.C. S 5108(c)(3); 28 U.S.C. 
SS 1346(a)(2), 1491(a)(2), 2401(a), 2414, 2510, 2517; 31 
U.S.C. S 1304(a)(3)(C) (1982); enacted November 1, 1978 by 
Pub. L. No. 95-563, 92 Stat. 2383. 

This report does not include an independent study of the fact 
of or causes of delay in processing contract disputes beyond 
the contracting officer level. However, the problem of 
substantial delay in litigating appeals before agency boards 
of concract appeals is well known by those who observe the 
process. Report of the Federal Contract Claims and Remedies 
Committee on Ways of Expediting Appeals Before the Boards of 
Contract Appeals . 16 Public Contract L.J., 161 (1986). 

Both caseloads and litigation time have increased sharply in 
recent years. For example, the number of docketed appeals at 
the Armed Services Board of Contract Appeals (ASBCA) 
increased 200% from 1978 to 1987. ASBCA's Report on Board 
Procedures During Fiscal 1986. 47 Fed. Cont. Rep. 93-95 
(1987). The growth in the number, complexity and cost of 
litigating these cases is further described in Crowell & Pou, 
Appealing Government Contract Decisions: Reducing the Cost 

[Footnote continued on next page] 



152 RICHARD J. BEDNAR 



the Government, failure to settle may mean litigating small claims 
whose litigation costs exceed the amount claimed. For the 
Government contractor, the CDA options that remain when settlement 
of the issues with the CO fails to materialize — litigate or 
forget it — simply are not satisfactory in many situations. To 
litigate a contract dispute before an agency board of contract 
appeals or in the Claims Court is to embark on a resource-costly, 
time-consuming venture. ^ To give up a claim thought to be 
valid — because it has been denied by the CO — may be a bitter, 
costly business decision. 

If the CDA options for resolution often are unsatisfactory, 
why then don't the parties caught up in Government contract dis- 
putes now make greater use of ADR technicjpies to settle their dif- 
ferences? The answer to that question is fairly complex. It 
depends on: (1) a general understanding of the primary Government 



[Footnote continued from previous page] 

and Delay of Procurement Litigation, 1987 Recommendations & 

Reports of the Administrative Conference (1988) 

(hereinafter Crowell & Pou) . The authors state that in 
recent years, 

. . . hearings on the merits became 
more formalized and extensive. Caseloads 
and backlogs increased, disputes became 
more heavily lawyered, and discovery and 
motions practice were introduced and ex- 
panded. More and more decisions took 
longer to be reached, and read. Many 
applauded these trends as enhancing con- 
tractors' due process rights; other view- 
ers decried them as inducing delay, 
bureaucratic irresponsibility, and liti- 
gation expenses . 

Whatever the merits of these various 
viewpoints, in 1978 the judicialized 
model of clairas resolution prevailed with 
the enactment of the Contract Disputes 
Act ( "CDA" ) , due in large part to a few 
court decisions finding broad due process 
rights and agitation by some private bar 
and board members . 

The procedures of the agency boards of contract appeals in 
particular allow for shortened processes in specific circum- 
stances. For excunple, decisions on motion for summary judg- 
ment, submissions on the record only, and speedier processing 
of smaller claims. See Arnavas, Donald P., Five Ways to the 
Merits; Some Suggestions on Seeking Expeditious Resolution 
of Contract Dispute Litigation . 19 Nat'l Cont. Mgmt. J. 11 
(Winter, 1985). 



ADR IN CONTRACT DISPUTES 153 



contracts disputes process; (2) an appreciation of the pivotal 
role of the Government contracting officer and the other principal 
participants in contract disputes; (3) a familiarity with the 
veurious AOR techniques that have potential application to this 
process; (4) an understanding of the key obstacles that lie in the 
way of greater ADR use; (5) a recognition of those conditions that 
are conducive to expanded use of ADR techniques; and finally, (6) 
an understanding of what specific actions are required to attain 
the objective of greater CO use of ADR techniques. This paper 
strives to support the proposition presented by addressing each of 
these particular considerations. 

The central focus of this report is on the Government CO who 
holds statutory authority to decide disputes that arise under or 
relate to a Government contract.^ Implicit in the CO's authority 
to decide disputes is the authority to resolve disputes by agree- 
ment with the claimant.^ Pursuant to the CDA, the CO decides 
which Government claims eure meritorious and should be presented to 
the contractor for payment. The CO also finally decides which 
contractor claims against the Government are compenseible and which 
are not. The finality of the CO's decision to deny a contractor's 
claims can be overturned only upon successful contractor appeal to 
the agency board of contract appeals or by successful direct ac- 
cess suit in the Claims Court. ^ The CO has continuing opportunity 
and authority to settle the disputed matter, extending from the 
time a contract controversy first arises until the matter is 
ultimately decided by a board of contract appeals or the Claims 
Court. ^0 In a real sense, the CO presides over the control valve 
to the whole disputes resolution process. If more disputes are to 
be settled the CO must be involved. Thus, this CDA process itself 
will be facilitated by encouraging COs to learn alternate dispute 



7 
8 

9 

10 



A detailed description of the role of the contracting officer 
in the Government procurement process is presented in Bednar 
& Jones, "The DOD Contracting Officer: A Study of the Past; 
An Assessment of the Present; Recommendations for the 
Future", A Report by the Ad Hoc Committee on the Role of DOD 
Contracting Officers, American Bar Association, Section of 
Public Contract Law (1987) (hereinafter The DOD Contracting 
Officer ) . 

41 U.S.C. S 605(a). 

Gene Peters v. United States, 694 F.2d 687 (Fed. Cir. 1982); 
Cannon Construction Co. v. United States, 162 Ct. CI. 94, 319 
F.2d 173 (1963). 

41 U.S.C. S 605(b). 

See Appendix A, Disputes Process , which shows the multiple 
opportunities for negotiated settlement in the disputes 
process. 



154 RICHARD J. BEDNAR 



resolution techniques and to use them in appropriate cir- 
cumstances . 

The principal objective of greater CO use of ADR in this con- 
text is to provide increased opportunities for fair, relatively 
inexpensive and relatively expeditious resolution of Government 
contract disputes by the contracting parties. ^^ By recognizing 
the usefulness of ADR in this decisionmaking process, by encour- 
aging greater ADR application, and by improving the ADR skills of 
those agency personnel who participate in the CDA process. Govern- 
ment contracting agencies can create a climate in which more dis- 
putes are rationally and justly settled without resort to litiga- 
tion. 

What is proposed in this study is not an alternative system 
to the CDA process, but rather some modest stimuli and incentives 
to improve the effectiveness of the CDA process itself. ^2 what is 
specifically suggested is expanded use of consensual forms of 
ADR — not ADR by regulatory mandate. Agencies as well as COs 
should be authorized and encouraged, rather than required, to 
expand the use of ADR methods at the CO level. 



il 



12 



Of course, th<=! agency boards of contract appeals were created 
for a nearly identical purpose — to provide informal, 
expeditious procedures for resolving disputes. See . 
Statement of Eldon H. Crowell before the U.S. Senate, 
Committee on the Judiciary, Subcommittee on Courts and 
Administrative Practice, Hearings on Administrative Dispute 
Resolution Act of 1988, S.2274, April 29, 1988. 

The policy of encouraging agencies and COs to explore the use 
of ADR in resolving matters between the agency and the con- 
tractor before the CO issues a final decision and after the 
resultant appeal is still in dispute has been supported by 
the Office of Federal Procurement Policy. Letter to ACUS 
from Robert P. Bedell, Administrator, OFPP, Dec. 10, 1987. 



ADR IN CONTRACT DISPUTES 155 



THE GOVERNMENT CONTRACTS DISPUTES PROCESS 

The first major event of the CDA disputes procedure is to 
direct a particular contractor or Government claim^^ to the con- 
tracting officer for a decision. The CO literally controls the 
sole point of entry to the contract disputes process.^** The CDA 
requires that all contract claims be presented in writing to the 
CO. It also requires that the CO issue a decision in writing and 
state the reasons for the decision. The CO's decision on contract 
claims is final and conclusive cind is not subject to review by any 
forum or agency, unless a timely appeal or suit is taken as pro- 
vided in the CDA. The contractor can litigate final CO decisions 
in either of two forums — the appropriate agency board of con- 
tract appeals ^5 or the Claims Court. ^^ Because the CDA greatly 
broadened the types of claims covered by the process, and 
specifically includes Government claims against the contractor, 
the vast majority of Government contract dispute resolution is 
accomplished in accordance with this CDA procedure. ^^ 



13 



14 



15 



16 



17 



A contract claim in this context is a written demand by one 
of the contracting parties (certified as required by the CDA) 
seeking as a matter of right, the payment of money in a sum 
certain, the adjustment or interpretation of contract terms, 
or other relief arising under or relating to the contract FAR 

5 33.201. 

This is not to suggest that the only ADR opportunities in the 
disputes process are at the CO level. For discussion of ADR 
opportunities beyond the level of the CO, see Crowell & Pou, 
Appealing Government Contract Decisions ; Reducing The Cost 

And Delay of Procurement Litigation . 1987 ACUS (1987). In 

addition to the specific ADR techniques discussed in Crowell 

6 Pou, the agency boards of contract appeals under the CDA 
have procedures intended to truncate the litigation process. 
The CDA requires these boards to provide for accelerated dis- 
position of claims of $50,000 or less and to establish both 
expedited disposition and simplified procedures for disputes 
of $10,000 or less. 

Currently, there are 12 federal agency boards of contract 
appeals, of which the Armed Services Board of Contract 
Appeals is the oldest and by far the largest. 

The option lies with the contractor. For discussion of 
factors which should be considered in making this binding 
election, see discussion in Cibinic and Nash, at 947-952. 

A discussion of the types of controversies subject to the 
disputes process established by the CDA is presented in 
Cibinic and Nash, Administration of Government Contracts, 
Second Edition, Government Contracts Progreun, The George 

[Footnote continued on next page] 



156 RICHARD J. BEDNAR 



To appreciate the prominent position of the CO in the dis- 
putes process, it is helpful to know the nature and sweep of the 
GO'S decisionmaking authority concerning contracts administration. 
FArIS S 2.101 defines a contracting officer as "a person with the 
authority to enter into, administer, and/or terminate contracts 
and make related determinations and findings ..." The text fur- 
ther explains that a single contracting officer may be responsible 
for duties in any or all of these areas. However, it is quite 
common, especially for larger contracting agencies and for major 
acquisition programs, to divide these basic duties cunong several 
contracting officers. For example, one contracting officer may be 
responsible for awarding contracts and modifications to contracts, 
another responsible for administering contracts, and a third re- 
sponsible for settling terminated contracts. ^^ Of course, 
contracting officers have supervisors, such as branch chiefs, 
office chiefs and heads of contracting activities. 

What the quoted definition of a CO suggests but does not make 
explicit is that this is a powerful Government official, with 
authority over the contract that is unlike that found in the pri- 
vate sector. 20 por exctmple, contract clauses such as Changes, 21 
Inspection . 22 g^d Termination for Convenience 23 give the CO the 
unilateral authority to alter significantly the bargain between 
the parties . 

Under the Changes clause, the CO may make unilateral changes, 
in specified contract features, within the general scope of the 
contract. The contractor must, in order to avoid breaching this 
material contract provision, continue performance of the work as 
changed, even if the contractor disagrees with the change. Then, 



[Footnote continued from previous page] 

Washington University, 1985 pp. 894-912 (hereinafter Cibinic 
and Nash) . 



18 

19 
20 

21 
22 
23 



"FAR" Stands for Federal Acquisition Regulation. The FAR was 
developed pursuant to the Office of Federal Procurement 
Policy Art, 41 US.C. SS 4 03(4) and 405(a), and is published 
in 48'C.F.R. . 

The POD Contracting Officer , at 26-31. 

In remarks at the 1987 Government Contractor Conference on 
January 29, 1987, Brig. Gen. Norman R. Thorpe, U.S. Air Force 
reported that the FAR uses the words "the contracting officer 
shall" a total of 1,879 times. 

FAR S 52.243-1. 

FAR S 52.246-2. 

FAR S 52.249-2. 



ADR IN CONTRACT DISPUTES 157 



unless the parties are able to agree on a contract price adjust- 
ment resulting from the change, the CO will unilaterally determine 
the amount, siibject to contractor remedies under the Disputes 
clause. 24 

The Inspection clause grants broad authority to the CO as the 
Government representative to assure that the contractor 
accomplishes all inspection and testing necessary to determine 
thi-t the supplies or services conform to the contract 
requirements. The clause requires the contractor to provide an 
inspection system acceptable to the Government. It also 
authorizes the Government to make inspections and tests while the 
work is in process. As the Government's representative, the CO 
may, as authorized by the Inspection clause, reject nonconforming 
items and services, require correction of defects, and even termi- 
nate the contract in circumstances where the contractor fails or 
refuses to comply. Disputes between the parties as to this 
inspection process are decided pursuant to the Disputes clause. 

The Termination for Convenience clause empowers the CO to 
terminate contracts for the "convenience of the Government", if 
the CO determines that a termination is in the Government's inter- 
est. The clause gives the Government the broad right to terminate 
without cause, and to limit the contractor's recovery to incurred 
costs, profit only on work accomplished, and costs of preparing 
the termination settlement proposal . There is no recovery for 
anticipated contract prof its. 25 Administration of this clause is 
also subject to the Disputes clause. 

These three clauses — Changes . Inspection and Termination 
for Convenience — are cited merely as examples of the unique 
authority held by Government COs. This CO authority is unlike 
that found in the commercial sector. 

Administration of these clauses spawns disputes, but these 
examples are by no means exhaustive of the class. Other examples 
of typical contract disputes would include cost or pricing issues, 
specification ambiguities and other contract interpretation con- 
troversies. Other contract clauses and other contract adiainistra- 
tion issues place the CO squarely in the role of decisionmaker for 
the Government. These Government contract clauses require 
decisions in the context of the CO acting as the contracting agent 



24 

FAR S 52.233-1. The Supreme Court has recognized that this 

power to direct changes j.s an important right in a Government 

contract. Choteau v. U.S., 95 U.S. 61 (1877). Also see 

vonBaur, The Origins of the Changes Clause in Naval 

Procurement . 8 Pub, Cont . L.J. 175 (1976). 

25 

The courts have recognized this unilateral power, 

unparalleled in the commercial section, as an important 

element of public policy. G.L. Christian and Sons v. U.S., 

160 Ct. CI. 1; 312 F.2d 418 (1963). 



158 RICHARD J. BEDNAR 



of the Government and directly chargeable with protecting the best 
interests of the United States . 

In an entirely different context, as mandated by the CDA, the 
CO is required to make decisions under the Disputes clause of the 
contract, not as the contracting representative of the United 
States, per se, but rather in the context of one standing apart 
from the controversy and acting as an impartial judge. 26 Here 
lies the opportunity for expanded use of ADR techniques. This 
vital decisionmaking function should not be constricted by old 
habits and rote practices. It is in this role of deciding 
contract disputes that COs should be authorized and encouraged to 
use whatever disputes resolution technique is best suited for the 
circumstances. It is in this context that federal contracting 
agencies27 ^j-q missing opportunities to enjoy the resource savings 
of litigation avoidance. It is in this context that there exists 
opportunity for greater use of ADR methods. 28 



26 



27 



28 



Quite unlike the day-to-day duties of the CO to vigorously 
protect the interests of the United states under the con- 
tract, the CO is required to be unbiased and impartial in 
deciding the merits of a contractor's claim. In Penn Instal- 
lation Corp. V. United States, 116 Ct. CI. 550, 89 F. Supp. 
545, aff 'd . 340 U.S. 898 (1950), the Court of Claims observed 
at page 563: "In other words, the contracting officer must 
act impartially in settling disputes. He must not act as a 
representative of one of the contracting parties, but also as 
an impartial, unbiased judge." 

The law does not permit the contracting officer to blindly 
follow the advice of technical advisors. For example, in 
Shirley Contracting Corp., ASBCA No, 29848, 87-2 BCA 
1 19,759, the Board found the final CO decision was flawed 
because the CO accepted an audit report which was obviously 
inadequate and superficial. More discussion of this legal 
requirement is presented later in this report. 

Alternate dispute resolution techniques have been widely used 
in the private sector and by state and local jurisdiction. 
The federal government has been notably slow in picking up on 
these techniques. Smith, Alternate Means of Dispute 
Resolution; Practice and Possibilities in the Federal 
Government . Mo. J. of Dispute Resolution 9 (1984). Also see 
Dispute Resolution . 88 Yale L. J., 905, 905-1104 (1979). 

The U.S. Army Corps of Engineers was the first federal agency 
to establish an ADR program. ADR Report . June 23, 1988, Vol. 
2, p. 219. 



» 



ADR IN CONTRACT DISPUTES 159 



THE KEY ROLE OF THE 
CONTRACTING OFFICER IN RESOLVING DISPUTES 



As a body politic, the Government of the United States can 
act only through its agents. For most functions involving the 
formation and administration of contracts, that agent is the con- 
tracting officer. 29 But, it would be a mistake to conclude that 
thr CO acts alone in the formation and administration of Govern- 
ment contracts. Successful contracts administration in particular 
depends on daily interaction between Government personnel and con- 
tractor employees . Among the key Government participants in this 
interaction are auditors, inspectors, engineers, production spe- 
cialists, safety specialists, contracts administrators, and many 
others whose efforts are essential to efficient contract 
administration. ^^ 

With all of this daily dialogue and interaction between rep- 
resentatives of the contracting parties, controversies and dis- 
agreements are bound to occur. Most such controversies and dis- 
agreements which emerge during this process eventually are settled 
by agreement between the disputants and do not grow into a formal 
"claim" by one party against the other. -^^ It is common for these 
contracts administration issues to be amicably resolved without 
the direct participation of the CO. In fact, the disputes process 
would collapse of its own burden if all controversies involved the 
personal attention of the CO. Routine disagreements, often issues 
of contract interpretation, generally are resolved informally at 
the level at which they arise. Many of these first-tier 
controversies involve minor issues of a technical nature and are 
resolved through discussions between technical personnel on an ad 
hoc basis. The disputants reach an informal compromise, or one 
disputant acquiesces in view of the other, or the matter is simply 
forgotten. It would be rare for disputes resolution techniques 
more sophisticated than ad hoc negotiation to be helpful in these 
routine dispositions of controversies. 

However, those contract dispute issues that are elevated to 
the level of the CO present an entirely different opportunity. 
Disputes rising to this level invoke the CDA process. Tha CO and 
team of advisers must address the disputed issues as required by 
the CDA. It is here that ADR fits naturally into the process, as 
it is express Government policy to try to resolve these disputes 



29 

Shedd, Joel P., Principles on Authority of Contracting Offi- 
cers in Administration of Government Contracts . 5 Pub. Cont. 
L.J. 88 (1972). This article provides an excellent 
discussion of the source and extent of CO authority. 

The POD Contracting Officer , at 41-53. 

Cibinic and Nash, at 926. 



160 RICHARD J. BEDNAR 



32 
by mutual agreement. Realization o£ this policy objective 

depends on three principal factors : 

o The authority of the CO to deal with the 
particular matter. -^^ 

o The adecjuacy of the relevant information 
brought to the CO's attention. ^^ 

o The willingness of the parties to reach 
settlement agreement . 35 

FAR S 33.210 describes the broad settlement authority of the 
CO in these terms : 

"Except as provided in this section, contract- 
ing officers are authorized, within any spe- 
cific limitations of their warrants, to decide 
or settle all claims arising under or relating 
to a contract subject to the Act. This autho- 
rization does not extend to — 

(a) A claim or dispute for penal- 
ties or forfeitures prescribed by 
statute or regulation that another 



32 



33 



34 



35 



It has long been Government policy to try to reach a nego- 
tiated settlement with the contractor before turning to the 
formal disputes process. Cibinic and Nash, 927. In fact, it 
is specific Government policy to try to resolve all contract 
claims by mutual agreement at the CO level, without litiga- 
tion. Office of Federal Procurement Policy Regulation, 45 
Fed. Reg. 31035 (1980); FAR S 33.204. 

The authority of the CO is not boundless. For example, the 
CO has no authority to settle, compromise, pay or adjust any 
claim involving fraud. FAR § 33.210. For another example, 
the CO has no authority to settle a tort claim that is not 
connected with the contract. Cibinic and Nash, at 901-902. 

A classic complaint of Government COs is that they are unable 
to settle the contract dispute because the contractor has not 
furnished adequate facts to substantiate the claim. Faced 
with inadequate factual justification, the CO has no alterna- 
tive other than to deny the claim. Interview with Frank 
Carr, Chief Trial Attorney, U.S. Army Corps of Engineers, 
May 9, 1988. 

Willingness turns on several factors, including attitude, 
pressures or lack thereof from others, and finding the time 
to become involved. 



ADR IN CONTRACT DISPUTES 161 

Federal agency is specifically au- 
thorized to administer, settle, or 
determine; or 

(b) The settlement, compromise, 
payment, or adjustment of any claim 
involving fraud." 

These limitations on the CO's settlement authority derive 
from the statutory limitations contained in S 6(a) of the CDA. 
Requests for relief that have been held to be outside the settle- 
ment authority because they are "penalties or forfeitures" include 
disputes concerning Davis-Bacon Act^^ wage classifications and the 
Contract Work Hours and Safety Standards Act.^^ Where there is 
evidence of fraud, the CO refers the matter to the agency official 
responsible for fraud investigation. 38 

If the parties are unable to dispose of a dispute through 
settlement, the disputes process contemplates the issuance of a 
contracting officer's "final" decision. A CO's final decision is 
not appealable by the Government, -^^ and the contractor seeking 
further consideration of the dispute must appeal or file suit 
within the timeframe and in the manner provided by law. 

The courts and the various boards of contract appeals have 
construed the standard Disputes clause as requiring a contracting 
officer to make a "personal and independent" judgment on the mer- 
its of a contractor's claim or Government claim. This requirement 
has been held to preclude a CO from agreeing to submit a claim to 
binding arbitration. In Dames & Moore . IBCA 1308-10-79, 81-2 BCA 
1 15,418, the Department of Interior Board of Contract Appeals 
said that an agreement to arbitrate cannot relieve the CO of the 
duty to make a personal and independent decision on contract dis- 
putes, or relieve the Government and the contractor of their duty 
to resolve disputes in accordance with the Disputes clause. This 
decision, as well as the position of the Comptroller General, 
discussed later in this report, have effectively stopped use of 



36 

37 
38 
39 



40 U.S.C. S 276. 

40 U.S.C. S 327. 
FAR S 33.209. 

41 U.S.C. S 605(b). The question may be asked, if the CO is 
truly impartial and unbiased in rendering a final decision, 
why shouldn't the Government have the same appeal 
opportunities as the contractor. As discussed in a later 
section of this report, the Government final decisionmaking 
process involves the interested elements of the Agency such 
that the CO's decision typically represents a consensus 
position. 



162 RICHARD J. BEDNAR 



binding arbitration as a method for federal contract disputes 
resolution. 

Under most circumstances, a contracting officer should and 
does seek legal and technical advice in arriving at a final deci- 
sion under the Disputes article.'*'^ The requirement for indepen- 
dence in the decisionmaking process is satisfied if the final 
decision is the contracting officer's ovm by adoption.**^ In fact, 
the "personal and independent" requirement may be met even though 
the decision is based principally upon legal advice given the CO 
by the Government attorneys. '2 

But, a decision by a contracting officer who follows the com- 
mand direction of a superior without making a personal and inde- 
pendent judgment on the merits of a contractor's claim has been 
held to be an invalid final decision under the Disputes clause. ^^ 
Instead, such situation may well be determined to be an abdication 
of CO responsibility. 

The contracting officer is required to make an unbiased and 
impartial determination of the merits of the contractor's claim. 
Certain agency regulations reqpiire contracting officers to refer 
proposed final decisions to superiors for review.'*^ It has been 
held that a CO's decision submitted to a superior for "approval" 
nevertheless may constitute a valid final decision. ^^ 



40 

41 

42 

43 
44 



45 



FAR S 33.211(a); FAR S 1.602-2(c). 

Max Jordan Bauunternehmung , ASBCA No. 23055, 82-1 BCA 
1 15,685 (1982). 

Pacific Architects & Engineers, Inc. v. United States, 203 
Ct. CI. 499, 491 F.2d 734 (1974). Also see Cibinic and Nash, 
at 941. 

See cases discussed in Cibinic and Nash, at 942, 943. 

The internal review process does not involve the participa- 
tion of the contractor. However, the bill which eventually 
became the CDA included up to the end of the legislation 
process a provision granting the contractor the right to an 
informal conference above the level of the CO. The purpose 
of such informal conference was to encourage settlement of 
the claim. This feature was stricken from the bill because 
Admiral Rickover and others thought it would undermine the 
authority of the contracting officer. 124 Cong. Rec. 36267. 

Jacob Schlesinger, Inc. v. U.S., 94 Ct. CI. 289 (1941); 
Penner Installation Corp. v. U.S., 116 Ct. CI. 550, 89 
F. Supp. 545, aff 'd . 340 U.S. 898 (1950). 



ADR IN CONTRACT DISPUTES 163 



A contracting officer's final decision becomes binding on 
both parties unless a timely appeal is taken."*" Under the CDA, a 
final decision becomes "final and conclusive and not subject to 
review by any forvun, tribunal or Government agency" unless ap- 
pealed to a board within ninety days after receipt of the deci- 
sion, or to the Claims Court within 12 months following its 
receipt .^^ 

FAR Subpart 33.2, which implements the CDA, establishes addi- 
tional policy and procedure for deciding Government contract dis- 
putes. The Government policy encouraging resolution all 
contractual issues by mutual agreement at the CO level, without 
litigation, is clearly expressed in FAR S 33.204. In an apparent 
gesture to ADR, this section includes the following suggestion: 

In appropriate circumstances, the contracting 
officer, before issuing a decision on a claim, 
should consider the use of informal discus- 
sions between the parties by individuals who 
have not participated substantially in the 
matter in dispute, to aid in resolving the 
differences. 

This suggestion for a "fresh look" at the issues recognizes the 
potential usefulness of an objective evaluation of the dispute 
from the perspective of those not directly involved in creating or 
perpetuating the dispute. This simple suggestion is predicated on 
the fact that sometimes particular individuals can be involved so 
deeply in the disputed issues that their positions become 
nonnegotiable and unchangeable. ^^ The FAR suggestion also 
recognizes that settlement possibilities may be improved by having 
different representatives of the contracting parties weigh the 
issues. What is not said is how this process is to flow and how 
the results of this "fresh look" relate to the ultimate CO deci- 
sion. While the door is open to a variety of ADR possibilities, 
in actual practice ADR is rarely even considered. 

Beyond this modest, oblique reference to one form of ADR, the 
FAR provides some specific guidance for application in the CO's 
contract disputes decisionmaking process. ''AR S 33.211 requires 
the CO to: 



In U.S. V. Holpuch, 328 U.S. 234 (1946), the court held that 
a contractor's claim was "outlawed" because the contractor 
did not appeal to the board within the time prescribed in the 
Disputes clause. 



47 
48 



41 U.S.C. S 609(a) . 

One of the reasons that cases may be settled even after 
getting into the grips of the litigators is that the 
litigators provide a now and detached assessment of the case. 



164 RICHARD J. BEDNAR 



o review the facts pertinent to the claim; 

o secure assistance from legal and other 
advisers ; 

o coordinate with the contract administration 
officer and others as appropriate; and 

o prepare a written decision. 

The issuance of a final decision and the filing of an appeal 
by the contractor does not end the contracting officer's authority 
to settle the claim. ^^ If a contracting officer enters into set- 
tlement negotiations after issuing a final decision but before the 
time for appeal has expired, the decision can lose its 
"finality, "^0 

In fact, the CO retains settlement authority even during lit- 
igation at the board of contract appeals. 51 For example, settle- 
ment agreements reached after a decision by the Board but prior to 
the expiration of time for a motion for reconsideration have been 
held to constitute an accord and satisfaction binding both par- 
ties. 52 

However, the CO's authority to settle claims does not extend 
to cases where litigation has commenced in a court, because fed- 
eral law grants the Attorney General sole authority to settle 
cases being litigated in the courts. 53 Nevertheless, the CO can 
and usually does strongly influence the outcome by assuring the 
CO's position is known and understood. 

The rather "bare bones" FAR guidance to COs results in con- 
sidereible variety in the specific way agency contracting officers 
actually decide contract disputes. The following are examples of 
the variations eunongst agencies: 



49 
50 
51 

52 
53 



See chart at Appendix A. 

Cibinic and Nash, at 929. 

The CO has authority to settle until the rights are vested. 
American Bosch Arma Corp., ASBCA No. 10305, 67-2 BCA 1 6564 
(1967) . 

Cibinic and Nash, at 930. 

28 U.S.C. SS 516 & 519; Executive Order No. 6166; United 
States V. Newport News Shipbuilding & Dry Dock Co., 571 F.2d 
1283 (4th Cir.), cert , denied . 439 U.S. 875 (1978). 



54 



55 



56 



57 



58 



ADR IN CONTRACT DISPUTES 165 



o The CO makes the final decision after draw- 
ing in advice and information from others 
as necesseury.^^ 

o The CO is required to coordinate or confer 
with a superior agency official or commit- 
tee for approval before issuing a final 
decision. ^^ 

o The CO is required to obtain the review and 
comments of the Chief Trial Attorney before 
issuing a final decision. 5° 

o The CO is the nominal and apparent author 
of the final decision, but in fact the 
actual decisionmaker is someone else in the 
agency. 5^ 

o The CO is a member of the decisionmaking 
board or committee. 58 



This is the situation within the Environmental Protection 
Agency. COs are required to have final decisions reviewed by 
legal counsel and the Chief, Contracting Office, prior to its 
issuance. 48 C.F.R. S 1533.211. Many other Government 
agencies follow this seune procedure. 

The procedure in the Naval Facilities Engineering Command is 
to control authority to issue final decisions at a high 
level. In contractor claims less than $250,000, the final 
decision is issued by the Engineering Field Division; over 
$250,000 by the Naval Facilities Engineering Command. In 
both situations the policy is to try to settle the dispute 
before it rises to the level of a final CO decision. Naval 
Facilities Engineering Command Contracting Manual (NAVFAC 
p. 68) (Oct. 1987). 

This is the situation for the U.S. Air Force. AF FAR Supp. 
33.211(a)(2). The Chief Trial Attorney leads the office to 
which are assigned the attorneys who represent the Air Force 
in disputes before the Armed Services Board of Contract 
Appeals. 

This has been observed personally by the author in his 
experience with the disputed process. For example, where the 
dispute turns on a technical issue beyond the full 
understanding of the CO, the CO is likely to adopt the 
deteinnination of a technical advisor. 

This formal structure for advice would draw into conference 
all those advisors having special knowledge or experience 
related to the issues in dispute. The CO may or may not be 

[Footnote continued on next page] 



166 RICHARD J. BEDNAR 



What actual practice shows is that CO disputes resolution is 
very much a shared activity, with other Government persons partic- 
ipating in the process. In all situations the CO acts in concert 
with others and the difference is only in the extent of the 
structure and the formality of the process. Accordingly, for ADR 
to flourish, the active support of all who share in the disputes 
resolution process is important. 



[Footnote continued from previous page] 

the chairperson, but would come away with a decision devel- 
oped during the meeting of the board or committee. 



ADR IN CONTRACT DISPUTES 167 



CO RELATIONSHIPS WITH 
OTHER GOVERNMENT PARTICIPANTS 

Only in rare instances would the CO act alone in deciding a 
Government contract dispute. 59 Depending on the nature of the 
problem, sp>ecialists in pricing, auditing, production, packaging, 
maintenance, quality control, transportation, contract administra- 
tion, contract law and various fields of engineering are available 
from within the agency or within the Government . ^0 j^ most con- 
tract dispute situations several Government advisers participate 
directly or indirectly. Perhaps the most common participants are 
attorneys , auditors , engineers and program managers . ^ ^ The CO 
interacts with these professional and technical specialists to 
obtain advice, information and direction. Each of these partici- 
pants in the disputes process, and others not mentioned as exam- 
ples, has a strong influence on the disposition of issues. ^2 They 
are positioned to encourage or discourage wider use of ADR tech- 
niques . 

1. Attorneys . 

The CO is required by FAR SS 1.602-2(c) and 33.211(a)(2) to 
obtain legal advice in the process of developing a decision on a 
contract dispute. Government attorneys who provide that advice 
ordinarily are knowledgeable and experienced in Government 
contract law and disputes procedure. Most attorneys are assigned 
to legal offices reporting directly to senior agency management .^^ 
In this organizational structure, neither the CO nor the attorney 



59 



60 



61 



62 



63 



In most dispute situations it would be very poor judgment for 
the CO to act without first gaining legal and technical 
advice. Barringer & Botke, IBCA No. 428-3-64, 65-1 BCA 
1 4797 (1965) . 

See Briefing Paper No. 86-4 (Reidl & Bastianelli) "Con- 
tracting Officer Authority", Federal Publication.*', Inc., 
March, iy86. 

For a fuller discussion of these relationships with the CO 
see The POD Contracting Officer , at 41-53. 

An experienced Government attorney wrote, "Often, the dispute 
is between the contractor and the Government engineer, in- 
spector, auditor, or disbursing officer. I have seen numer- 
ous cases in which the CO sided with the contractor against 
these Government officials." Letter to the author from James 
F. Nagle, Attorney, Contract Law Division, Department of the 
Army, March 28, 1988. 

The DoD Contracting Officer , at 46. 



168 RICHARD J. BEDNAR 



is the superior or subordinate of the other. The Government at- 
torney is not a CO and has no independent authority to bind the 
Government in contract.^'* Nevertheless, because of the attorney's 
professional stature and lofty position within the agency 
organization, the attorney has significant influence over CO deci- 
sions. In some situations, the legal adviser to the CO may also 
be the attorney to represent the agency before a board of contract 
appeals if the contractor appeals an adverse final decision. In 
practice, the CO tends to be deferential to the advice of the at- 
torney and acts contrary to that advice only in unusual circum- 
stances. 

Thus, Government attorneys could stimulate and influence 
expanded use of ADR in settling contract disputes. ^5 with some 
exceptions, this has not been the prevailing condition. The rea- 
sons why most Government attorneys have not been more forceful in 
promoting ADR are not objectively demonstrable. There are some 
likely reasons. All attorneys tend to be conservative in doing 
their work and stick to conventional approaches to solving 
problems. Government attorneys are particularly familiar with the 
CDA process and are comfortable in working within that traditional 
method of disputes resolution. Many are unfamiliar with ADR and 
would be uncomfortable recommending ADR use. On the other hand, 
the job of providing legal advice to the CO could extend to 
suggesting that a particular case was appropriate for ADR. While 
it is for Government managers rather than their lawyers to give 
official concern to conserving agency time and money resources and 
to seek more efficient ways to do the agency's work. Government 
lawyers do have a strong interest in efficient legal methods. 
Many Government attorneys surely have felt the frustrations of the 
CDA disputes process and could be strong advocates of ADR once 
convinced of its benefits to the process. ^^ Significant 
acceptance of ADR may come once more Government attorneys 



64 
65 



66 



Defoe Shipbuilding Co., ASBCA No. 17095, 74-1 BCA 1 10,537 
(1974). 

Interview of Lester Bdelman, Chief Counsel, U.S. Army Corps 
of Engineers, April 8, 1988. The Corps of Engineers has set- 
tled several contract disputes pursuant to ADR techniques, 
and is recognized as the lead federal agency in this regard. 
As attorneys have strong influence within the organization, 
it is important to train them in ADR techniques . 

Marquerite Millhauser, in her essay "The Unspoken Resistance 
to Alternate Dispute Resolution," Negotiation Journal , Janu- 
ary 1987, suggests that, because of self-doubts or skepti- 
cism, a lawyer's first reaction to a proposal to use ADR is 
likely to be aggressive opposition. But with calm persis- 
tence by the side first proposing ADR, the lawyer on the 
other side may be more willing to accept the proposal on its 
face and embrace it as his or her own. 



ADR IN CONTRACT DISPUTES 169 



recognize that these techniques often provide a more efficient 
method of disposing of claims than does litigation. 

2. Auditors. 

COs also tend to be deferential to the conclusions of Govern- 
ment auditors. By regulation, if an agency CO disagrees with the 
auditor's conclusions in an audit of a contractor's pricing 
proposal, the CO is required to explain the basis for the 
disagreement in writing. ^^ Most COs axe strongly impelled by 
practice as well as by regulation to accede to the conclusions of 
the auditor. An auditor's advice ordinarily is rejected only 
after deliberate review of the auditor's advice. ^^ 

As auditors deal with precise numbers, it is arguable that 
auditors are by nature opposed to compromise settlement of dis- 
putes. On the other hand, few contract disputes turn exclusively 
on facts and figures which are objectively determinable. For 
example, while it may be objectively determinable how much expense 
a contractor has incurred on a given contractual requirement, the 
real issue in dispute may be how much of that cost incurrence was 
reasonable. For another excunple, it may easily be shown the 
amount of money originally bid on a job, when the disputed issue 
may be whether the job was underbid. Both of these examples 
provide room for judgment and compromise. In deciding a dispute, 
the CO would use the auditor's information but also would use 
judgment. In short, there is no reason inherent in the nature of 
their work for auditors to be opposed to wider use of ADR.^^ 
Auditors may be expected to raise a strong voice favoring a well- 
reasoned, auditable, written ADR settlement agreement reached 
through ADR methods. ^^ Having gained that assurance, auditors 
reasonably may be expected to join in support of expanded ADR use. 



^^ FAR 5 15.803(c) . 



68 

The POD Contracting Officer , at 47; DOD Directive 7640.2. 

69 

Interview with Michael J. Thibault and David Capitano, 

Defense Contract Audit Agency, May 13, 1988. 

70 

The Corps of Engineers used a minitrial to resolve a dispute 

on the Tennessee-Tom Bigbee waterway project. The con- 
tractor's claim of $60 million was settled for $17.25 
million. The DOD Inspector General investigated the 
settlement. The Inspector General found no fault with the 
use of a minitrial or with the settlement, but did criticize 
the adequacy of the supporting documentation for arriving at 
the amount due. Sourcebook, 581-586. 



170 RICHARD J. BEDNAR 



3. Engineers . 

COS are rarely prepared by education or experience to deal 
comfortably with the technical aspects of a dispute. When techni- 
cal issues are present the CO looks to Government engineers for 
technical advice. Government engineers prepare contract specifi- 
cations, evaluate contractor technical proposals and assist COs 
directly in evaluating claims. Engineers frequently are appointed 
as CO technical representatives for contract administration. 
Sometimes engineers have direct knowledge of the facts giving rise 
to the dispute. Sometimes the engineers have authored the 
specification about which the parties are in controversy. Thus, 
engineers play an important and sometimes decisive role in 
disputes resolution.'^ 

Engineers tend to have strong views as to the technical cor- 
rectness of a contract claim, but they also tend to be practical 
minded. Engineers may thus be expected to support wider use of 
ADR in disputes resolution, if they believe that ADR provides an 
efficient forum for full consideration of the technical aspects of 
the disputed matter. 

72 

4. Program Managers . 

Most Government program managers are found in DOD. A program 
manager has the responsibility for acquiring and fielding a major 
system. Thus, the progreim manager has centralized management 
authority over all technical and business aspects of major systems 
acquisition. In this sense, the program manager controls the 
funds, and speaks for the agency element which generated the 



71 



72 



The DOD Contracting Officer , at 44 . In disputes involving 
technical issues, the CO would look to the Government's 
technical experts to evaluate the facts and circumstances and 
to develop a position. That position ordinarily is adopted 
by the CO. Sometimes, during the course of negotiations with 
the contractor's representatives, the CO will be persuaded to 
use the contractor's technical position and will, in effect, 
overrule the Government's technical position. 

Much has been written about the relationship of the CO and 
the Program Manager. The Packard Commission report A Quest 
for Excellence . June 1986, includes the recommendation: 
"Establishing short, unambiguous lines of authority would 
streamline the acquisition process and cut through bureau- 
cratic red tape. By this means, the DoD should substantially 
reduce the number of acquisition personnel." 

In the report, The DoD Contracting Officer , it is recommended 
that program managers and COs train together. 



ADR IN CONTRACT DISPUTES 171 



requirement for the contract and is awaiting the completion of 
contract performance. While having overall authority and 
responsibility for the acquisition of a system, program managers 
generally do not have authority to enter contracts . Contractual 
arrangements are the province of the CO. The organizational 
relationships between the program manager and the CO tends to fall 
into one of four categories: 

o The CO works in the contracts department, 
but provides direct support to the program 
manager on a full or part-time basis. 

o The CO works in the program office, but 
repoirts to and is supervised by the con- 
tracts department . 

o The CO both works in and is supervised by 
the program manager. 

73 
o The program manager is appointed as a CO. 

Whatever the organizational relationship, there is a close tie 
between the CO and the program manager which is reflected also in 
their approach to disputes resolution. Very broadly speaking, the 
program manager drives to meet the program needs of the agency, 
while leaving the contract-related problems in the hands of the 
CO.^** As the focus of the program manager is on timely acquisi- 
tion and fielding of the system, contract disputes become matters 
of concern to the program manager in direct relationship to their 
potential program impact. If the dispute disrupts or does not 
materially impact the program, the typical program manager would 
tend to defer to others on how best to bring the matter to 
resolution. On the other hand, if the contract disputes or looms 
as a threat to the program, the program manager will be a key 
actor in seeking a prompt and satisfactory resolution. ^5 



73 



74 



75 



One reason why the Corps of Engineers has found success in 
using ADR methods in resolving contract disputes is that the 
commander (District Engineer) is appointed as a CO for the 
ADR process. 

A discussion of this relationship is presented in 
Higginbotham, Richard W. , Program Manager and Contracting 
Officer; What are Their Roles ?. Contract Management, April 
1987, at 12. 

In dealing with disputes arising out of contracts which are 
not under a program manager, the CO will contend with an 
agency official representing the ultimate user of the goods 
or services being required. The "user" also provides the 
funds for the acquisition and, accordingly, is another team 
person to deal with in the disputes resolution process . 



172 RICHARD J. BEDNAR 



Accordingly, there are no irreconcilable differences between 
the roles and interests of those who participate most directly in 
the disputes process and the expanded use of ADR. Disputes 
resolution should be viewed as a team effort, with a number of 
essential participants in direct support of the CO. Strengthening 
this team work through increased use of ADR techniques is 
consistent with best management practices and catches the spirit 
of the much respected Packard Conuoission Report for greater ef- 
ficiency in procurement.^^ Wider ADR use in contract disputes 
resolution is dependent in large part on gaining the enthusiastic 
support of those Government advisers and officials who hold major 
roles in the contracting process as well as in the more narrow 
disputes resolution process. This is a tough challenge, but a 
challenge that can be met with time, patience and persistence. 
Some modest suggestions for gaining agency support of greater ADR 
use in resolving contract disputes at the CO level are presented 



in a later section. 



77 



76 



77 



See Kavanaugh, Jack & Michelle Kalkowski, The Packard Commis- 
sion; A Blueprint for Change . Contract Management, April, 
1986, at 14. 

Participation of these advisors to the CO in the CDA process 
provides a sort of multiple check and balance. This term 
approach tends to protect the public interest that important 
decision be rationally made. For this additional reason, it 
is important that these s£une team members have a role in ADR 
methods. This has been the experience of the Corps of 
Engineers and the Navy in their successful use of ADR in 
resolving Government contract disputes . 



ADR IN CONTRACT DISPUTES 173 



KEY FORMS OF ADR 
SUITABLE FOR CONTRACT DISPUTES 

The law is clear that actual authority is required to bind 
the Government in contract. ^^ The CO is the Government official 
authorized to execute contractual documents which bind the United 
States. ^^ Similarly, only the CO holds authority to decide dis- 
putes arising under or relating to such contracts. Thus, there 
are serious practical and legal implications in pursuing any ADR 
approach which purports to work outside the CO's authority. Prop- 
erly structured, however, there are a number of ADR techniques 
which are consistent with the CO authority, which may even have 
the effect of strengthening that authority, and which are per- 
fectly suitable for resolving Government contract disputes . 

1 . Negotiation . 

Because ad hoc negotiations between the parties to the 
contract dispute already is a familiar technique under CDA 
procedures, one does not readily reflect on it as a form of ADR. 
However, negotiation is a key ADR method, and one which should be 
developed for greater use at the CO level . Skilled in negotiation 
techniques, COs could do much better in their bargaining, whether 
with contractors or informally with agency personnel. Many COs 
doubtless already see themselves as experienced and competent 
negotiators. However, experience has shown that increased train- 
ing and attention to listening and communication skills, use of 
"interest" and "principled" rather than "positional" bargaining, 
and similar negotiation approaches can significantly improve many 
persons' abilities. Present CO training does not specifically 
address these skills, and thus presents a gap in their skills and 
self-image as professionals.^^ 

As developed earlier, the CO also holds a key role as "team 
builder" of those who participate in the disputes resolution 
process. ADR, including in particular improved negotiation 
skills, can better enable a CO to facilitate meaningful discussion 
with a contractor by first working to develop a coherent intra- 
agency position that takes into account the views and needs of 
attorneys , auditors , program managers , engineers and others . 

In this way, even without the assistance of a third-party 
neutral, the CO can help avert appeals by reducing the number of 



Federal Crop Insurance Corp. v. Merrill, 332 U.S. 380 (1947) 
79 



78 

FAR S 2.101. 

80 

For discussion of key self perceptions of DoD contracting 

officers, see. The DoD Contracting Officer , at 82. 



174 RICHARD J. BEDNAR 



situations where it is easier simply to pass the buck to the board 
of contract appeals because of internal disagreements or incoher- 
ent positions. 

81 
2 . Arbitration . 

The essence of arbitration is that a neutral third party 
selected by the disputing parties decides the issues submitted 
after receiving evidence and argument from the parties. The neu- 
tral commonly is selected because of subject-matter expertise. 
Arbitration may be binding on the parties by operation of law or 
through contractual agreement. Non-binding arbitration also is 
practiced to a more limited extent. Arbitration also may be 
either voluntary or mandatory, depending on the basis of the 
structure. Arbitration is widely used in resolving disputes 
between Government prime contractors and subcontractors . 82 it is 
also widely used in the commercial sector, especially in labor- 
management disputes. °3 Arbitration may be structured to permit a 
compromise solution. 

In the Government contracts context, as in many others, the 
Comptroller General has held that the Government cannot be bound 
by arbitration unless the agency has specific statutory authoriza- 
tion, or the arbitrator is limited to fact finding. 8*. 



81 



82 



83 



84 



An excellent discussion of arbitration in the Government con- 
tracts context is presented in Crowell & Pou, 54-57. Also 
see Elkouri, Frank & Edna A. Elkouri, How Arbitration Works , 
D.C.: BNA, 1960. 

Subcontractors generally do not have the right to take 
contract claims directly with the Government because of the 
lack of privity. FAR S 44.203(b)(3). 

Hardy, Timothy S. and R. Mason Cargill, Resolving Government 
Cont ract Disputes; Why Not Arbitrate? . 34 Fed. B. J., 20 
(1975) . 

The Administrative Conference and others have called for 
Congress to overturn this view. Administrative Conference 
86-3, 1 C.F.R. S 305.86-3; Harter, 1986 Recommendations and 
Reports of the Administrative Conference of the U.S. ( here- 
inafter 1986 ACUS); ADR Sourcebook, p. 309 (1987); Hardy & 
Cargill, Resolving Government Contracts Disputes; Why Not 
Arbitrate ?. 34 Fed. B. J. 1 (1975), ADR Sourcebook 351; 
Behre , Arbitration; A Permissible or Desirable Method for 
Resolving Disputes Involving Federal Acquisition and Assis- 
tance Contracts ?. 16 Pub. Cont. L. J. 66 (1986). A recent 
report to the Administrative Conference by Richard K. Berg, 
Legal and Structural Obstacles to the Use of Alternative 

[Footnote continued on next page] 



ADR IN CONTRACT DISPUTES 175 



Because of the CDA process and the position of the Comptrol- 
ler General, arbitration is not one of the favored ADR techniques 
presently used to resolve disputes between the Government and its 
contracting partners. 85 should the legal cloud cast by the 
Comptroller General be removed, such as by legislation, the way 
would be clear to consider arbitration more fully as a potential 
ADR technique. 

Given the CO's CDA decisionmaking authority, arbitration 
could be a viable consideration in at least two contexts: one, in 
a non-binding scenario where the neutral arbitrator's decision 
would be advisory only to the CO; two, in a scenario designating a 
particularly recognized and respected Government official as the 
CO and "arbitrator. *• By agreeing to making a CO the arbitrator — 
that is, a CO who theretofore was not involved in the disputed 
issues — the contracting parties would, in effect, facilitate the 
CO's decisionmaking responsibility under the CDA through a 
structured arbitration. An immediate concern is whether a 
contractor would agree to arbitration in circumstances where a 
Government official was designated as a contracting officer and 
arbitrator. The answer of course depends on the facts and 
circumstances. However, holding in mind that the CDA now requires 
the contracting officer to be unbiased and impartial in making a 
final decision under the Disputes clause, and that the CO's 
decision is often made on the basis of facts informally developed, 
there may very well be situations where the parties would agree to 
making the CO the arbitrator. Unlike the current process for 
reaching a final decision, where no procedure is prescribed for 
presenting facts and argument to the CO, traditional arbitration 
procedure at least would lend itself to a more disciplined 
presentation of facts and arguments to the CO in the role of 
arbitrator. In this context, placing a CO in the role of 
arbitrator would not be remarkably different from the CO's current 
CDA role as a neutral, final decisionmaker. In fact, the 
arbitration process would better ecjuip the CO to render a better 
informed decision. 



[Footnote continued from previous page] 

Dispute Resolution in Federal Programs . (September 1987), 
argues persuasively that GAO's objections are without legal 
foundation and unpersuasive. 

85 

There are exceptions to this general rule. In his unpub- 
lished paper "Alternative Dispute Resolution Methods for Gov- 
ernment Contracting" June, 1988, Frank Carr reports on the 
successful use of non-binding arbitration in several separate 
cases involving Corps of Engineers contracts. 



176 RICHARD J. BEDNAR 



3. Factfinding . 

The central idea of this AOR method is to locate and desig- 
nate an individual or panel with special qualifications in the 
subject matter of the dispute, who then develops the facts and 
presents them in a report to the contracting parties. ^^ The par- 
ties then may negotiate a settlement, embark on another ADR path, 
or place the matter back onto the CDA track. Considering that 
most Government contract disputes are heavily fact-driven, and 
that highly technical issues often separate the parties, factfind- 
ing has strong potential application to the Government contract's 
disputes resolution process. 

Consider for example a dispute focusing on the issue of 
whether certain sophisticated electronics work did or did not con- 
form to the technical specification of the contract. Using the 
technique of factfinding, the CO and the contractor might be eible 
to accelerate dispute resolution by gaining the informed, factual 
analysis and advice of a recognized, neutral, electronics expert. 

4. Mini trial. 



To the limited extent that ADR so far has been used in the 
resolution of Government contract disputes, the minitrial has been 
the preferred approach. 87 The Corps of Engineers, the Navy, NASA 
and others have used the minitrial with well-publicized success. ^^ 
In essence, the minitrial is a structured settlement process based 
on densely concentrated presentations of facts and argximents to 
senior officials for each side authorized to settle the dispute. ^^ 
The minitrial occurs before a final CO decision. In most 
Government contracts mini trials to date, a neutral adviser has 
participated to provide opinions and advice and to otherwise 
facilitate resolution of the issues. The idea of the minitrial is 
that the principals for each side will come to agreement after 



86 

87 
88 

89 



Paths to Justice : haior Public Policy Issues of Dispute 
Resolution . Report of the Ad Hoc Panel on Dispute Resolution 
and Public Policy, National Institute for Dispute Resolution 
(Oct. 1983), ADR Sourcebook , at 45. 

Crowell & Pou, at 19. 

Edelman, Lester & Frank Carr, The Mini-Trial; An Alternative 
Dispute Resolution Procedure . 42 Arb. J. 7 (1987) ADR Source- 
book . 231. 

Nash, A New Technique for Settling Disputes — The Minitrial 
(1985) (unpublished article); Killian, Paul & David Mancini, 
Mini-Trails; Basic Principles & Guidelines . Construction 
Briefings, Fed. Pub. (March 1985). 



ADR IN CONTRACT DISPUTES 177 



having heard the presentations of fact and argument within the 
catalytic presence of the neutral adviser. The CO or the CO 
Supervisor who also is a CO could very well be the principal for 
the Government in a minitrial structured to resolve a Government 
contract dispute. Such direct CO participation is seen as being 
in harmony with the CDA. If the minitrial produces agreement, the 
dispute is resolved. If the minitrial fails to produce agreement, 
the matter would be ripe for final decision and follow-on CDA lit- 
igation. 

5. Mediation . 

This techniqpie simply involves the assistance of a neutral 
third party in negotiating an agreement between the contracting 
parties themselves. ^0 It is less formal and less structured than 
the minitrial. The mediator has no authority beyond that bestowed 
by the parties and is not asked to render a decision. The CO 
could represent the Government at mediation without prejudice to 
the CDA responsibilities . Mediation is seen as a step above ad 
hoc negotiations, as it recognizes the potential for an earlier 
and better settlement by drawing upon the knowledge and experience 
of a trusted neutral. The mediator's role can be as active as the 
parties themselves permit, ^^ thus keeping the parties in control 
of the process. The mediator can be a source of fresh approaches 
to problem resolution . a sounding board for settlement ideas , and 
am impartial adviser. ^2 The skills of a mediator can produce 
agreement even in circumstances where the positions of the negoti- 
ators seem to be immutably hardened. ^^ 



90 



91 



92 



93 



Folberg, Jay & Alison Taylor, Mediation; A Comprehensive 
Guide to Resolving Conflicts Without Litigation . San Fran- 
cisco: Jossey-Bass (19&h). 

A strong mediator can play a role similar to a settlement 
judge by proposing settlement terms and urging the parties to 
accept them. Letter to ACUS from Peter H. Kaskell, Senior 
Vice President, Center for Public Resources, Nov. 17, 1987. 

Having an experienced, mediator is key. Having a mediator 
knowledgeable in the subject matter of the dispute is a 
distinct advantage. Interview of Judge Edward Houry, 
Department of Agriculture Board of Contract Appeals, by Mr. 
Wallace Warfield, ACUS, March 2, 1988. 

See discussion in Crowell & Pou, at 49-51. 



178 RICHARD J. BEDNAR 



6 . Settlement Judges 

This ADR technique contemplates the use of a special "settle- 
ment" judge to encourage the parties to resolve the dispute with- 
out compromising the position of the parties in subsequent litiga- 
tions should settlement not be attained. The settlement judge 
ordinarily is not the trial judge later in those situations where 
settlement does not materialize. Hence, this process is readily 
distinguishable from the routine practice of many trial judges to 
encourage the parties during pretrial to narrow the issues or set- 
tle the dispute. 

The GO'S role in this particulcir ADR process is subject to 
considerable variation. The use of a settlement judge could 
prompt the CO to undertake a relook of the disputed issues even 
after having made a final decision denying the contractor's claim. 
The perceived advantage of this approach is the opportunity to 
gain authoritative advice as to how the case might be decided in 
the settlement judge's court or board of contract appeals. The 
apparent shortcoming of this ADR method is that the settlement 
judge's advice necessarily would be based on something less than 
that information which would emerge during actual litigation. 
Furthermore, a settlement judge is aware of the forxim's caseload 
and could be influenced by that knowledge. Nevertheless, the way 
should be cleared for COs to seek assistance from a settlement 
judge in suitedsle circvimstances. 

94 

7 . Disputes Resolution Boards 

Recently, the Corps of Engineers began using a new ADR method 
that the Corps refers to as a "dispute resolution panel." 
Although this method is new to the federal sector, it has been 
used by several state governments in large construction projects 
performed in the western states. Under this ADR method, the par- 
ties agree that contract disputes arising during construction may 
be voluntarily submitted to the panel for -^n opinion as soon as 
the disputes occur. 

Under the Corps of Engineers program, a disputes resolution 
panel consists of three private technical experts. The government 
and the contractor each select one member of the panel and the 
third is selected by agreement of those two members. The 
procedure provides for disputes to be submitted to the panel as 
disputes inevitably emerge during actual construction. The panel 
makes a nonbinding written recommendation to the contracting 
officer and the contractor. 



94 

This description of the Disputes Resolution Board is based on 

an unpublished paper by Frank Carr, titled "Alternate Dispute 

Resolution Methods for Government Contracting", June, 1988. 



ADR IN CONTRACT DISPUTES 179 



Unlike most ADR methods which address disputes reactively, 
the dispute resolution board attempts to resolve a dispute at its 
inception, before the Government and the contractor have begun to 
expend resources in support of the litigation. Further, easy ac- 
cessibility to the panel and the prompt resolution of disputes 
should cause a minimum of disruption of the construction work and 
maintain an amicable working relationship between the parties . 

At this writing, two Corps of Engineers Districts are in the 
process of implementing this ADR method by including a disputes 
resolution board provision in several contracts. So far, however, 
this ADR technique is not known to have been used in Government 
construction contracting. 

8. Summary Trial With Binding Decision 



This is a procedure pursuant to which the scheduling of the 
appeal is expedited and the disputants try their appeal informally 
before an Administrative Judge or panel of judges. A summary, 
"bench" decision generally will be issued upon conclusion of the 
trial; or a summary written decision will be issued no later than 
ten days following its conclusion. The parties must agree that 
decisions, imlings, and orders by the Board under this method, 
against the Government or Appellant, shall be final, conclusive, 
not appealable, and may not be set aside, except for fraud. All 
such decisions, rulings and orders will have no precedential 
value. The length of trial and the extent to which scheduling of 
the appeal is expedited will be tailored to the needs of each 
particular case. Pre-trial, trial, and post-trial procedxires and 
rules applicable to appeals generally will be modified or 
eliminated to expedite resolution of the appeal. 

Because of the public interest involved, it is important in 
all settlement methods that the Government side has before it 
sufficient factual information on which to base a rational 
decision. In this regard, it should be understood that dispute 
resolution through use of an alternate technique would 
nevertheless require that the contractor provide sufficient facts 
to the Government and to certify in the submission of those facts 
for claims over $50,000 that: "the claim is made in good faith, 
that the supporting data are accurate and complete to the best of 
his knowledge and belief, and that the amount requested accurately 
reflects the contract adjustment for which the contractor believes 
the government is liable." This certification, presently required 



180 RICHARD J. BEDNAR 



95 
by the CDA for claims over $50,000 should continue to be 



required in the ADR context, 



95 

41 U.S.C. S 605(c)(1). The essential value of such certifi- 
cate is to place liability on the contractor for misstate- 
ments or false claims. The CDA gives the Government addi- 
tional protection against false claims by entitling it to 
recover the amount of the unsupported part of the claim, plus 
the costs of reviewing that part of the claim which states an 
overcharge. 41 U.S.C. S 604. 



ADR IN CONTRACT DISPUTES 181 



KEY OBSTACLES TO GREATER ADR USE 

A convincing case can be made for the proposition that agree- 
ment between the parties directly involved is the best way to 
resolve disputes which arise during the performance of Government 
contracts. The parties to the dispute are best positioned to know 
the facts. They even may have participated in creation of the 
issues which remain in dispute. They are best positioned to know 
the program and other business implications of settlement — 
certainly better positioned than would be a board or court. They 
often are strongly motivated to compose their differences and get 
on with the work which is the subject of the contract. By prompt 
settlement of the issues there is a real oppoirtunity to better 
control the outcome and to save both time and litigation costs. 

If those obvious virtues of early settlement of disputes by 
the contracting parties involved directly are valid, it then fol- 
lows that the parties to the dispute ought to be encouraged and 
motivated to settle rather than to litigate, and ought to be 
facilitated in the disputes resolution process through appropriate 
use of a variety of techniques and procedures. Particularly at 
the level of the contracting officer, there should be expanded use 
of dispute resolution techniques like those previously discussed 
in this report. 

The term "at the level of the contracting officer" is meant 
at the level of the contracting officer having authority to make 
final decisions within the meaning of the Contract Disputes Act. 
Said another way, the CO who is given responsibility for disputes 
resolution is not necessarily the CO who awarded the contract or 
is principally responsible for its administration. The term also 
should be understood to encompass a relatively wide band of time, 
not precisely delimited, including at the front end of the process 
settlement of contract administration issues which have not 
ripened into formal claims, and extending in time to include 
continuing opportunities to settle cases that have moved to the 
hands of the litigators after the CO's final decision. In short, 
ADR has a legitimate place in the decisionmaking process at any 
time the CO has authority to settle the contract dispute. 

Given the proposition that ADR techniques could foster set- 
tlement of more disputes, it becomes important to identify and 
consider those key obstacles standing in the way. The following 
are seen as the key impediments: 

1. Unawareness . 

Notwithstanding the recent welling of attention to ADR in the 
Government contracts context. Government agencies traditionally 
have relied on negotiations by COs or litigation before agency 
boards of contract appeals to resolve contract disputes. This has 



182 RICHARD J. BEDNAR 



been the primary path to resolution of contract disputes. There 
remains widespread lack o£ enlightenment and misunderstanding of 
what other ADR techniques are and how they might fit into the 
contract disputes resolution process. ADR simply is not well 
known by those responsible for contracts disputes resolution — by 
either the contractor or by the Government. 

The principal reason for this incognizance seems to be agency 
inertia. Contract disputes resolution follows a familiar path, 
the path specifically authorized by statute, pointed to by 
regulations, described in a contract clause and routinized in 
practice. Contract disputes resolution was simply not regarded as 
fertile ground for ADR techniques until the relatively recent use 
of minitrials by the Corps of Engineers and occasional ADR 
experiments in a small number of other agencies. As ADR has not 
been employed extensively in the contracts disputes process, there 
is no broad exposure to its qualities and applications. Thus, 
this lack of awareness remains a serious obstacle. Much work 
needs to be done by proponents of ADR to teach what these tech- 
niques are and to warm up the potential participants to the idea 
of ADR. As more Government contracts cases are successfully set- 
tled through an ADR approach, the good news of ADR gradually will 
spread and the obstacle of unawareness will be abated. 

2 . Apprehensiveness . 

There seems to be no end to the fears associated with first 
use of ADR. ^6 Among the most commonly misgivings mentioned in the 
literature and by persons interviewed are (a) fear that the con- 
tractor will agree to ADR for the real purpose of learning more 
about the Government's case — or vice versa ; (b) nervousness by 
the contracting officer that the ADR result will be criticized by 
the agency's inspector general or by the Congress; (c) skittish- 
ness that the "boss" will disagree with the settlement result; (d) 
concern that the private bar somehow will manipulate ADR practice 
to its advantage and to the disadvantage of the disputants; (e) 
misgiving that ADR somehow will carve into the contracting 
officer's turf or author.! ty; and {£} coiicern that ADR will expose 
weaknesses in the case. There are other fears, but these few are 
most frequently mentioned. The obstacle of apprehensiveness, like 
that of ignorance, will be overcome as ADR success breeds addi- 
tional success. There is no rational basis for any of these 
fears. If the ADR process produces agreement, it is ar. agreement 
achieved by persons of authority who are well positioned to know 
all the facts and implications of the settlement; certainly they 



96 



Fear, particularly fear of being criticized by others such as 
DCAA, may well be the major obstacle to wider use of ADR 
techniques at the CO level. Letter to the author from 
Jeffrey B. Trattner, Director of Purchasing, The Perkin-Elmer 
Corporation, August 4, 1988. 



ADR IN CONTRACT DISPUTES 183 



are better positioned than a court or board. If the ADR process 
results in failure to agree, at least the parties will have 
leftrned more about the strengths and weaknesses of their own case 
as well as that of their contracting partner. Surely, even a 
"failed" ADR experience will help expedite the preparation for any 
litigation which follows. In this sense, the "risk" of pursuing 
ADR is minimum. 

3. Concern That ADR May Not Save Time Or Money . 

Some suggest a reluctance to try ADR techniques because they 
are not convinced that time or money actually will be saved. ^^ 
ADR, it is asserted, may turn out to be a sidetrack from the 
straighter, more familiar path to resolution by litigation. It is 
true that there are no precise gauges by which it may be objec- 
tively demonstrated how much time or money that ADR will save. On 
the other hand, to the extent that settlement of the dispute by 
the parties generally is a cjuicker process than litigation, this 
is a baseless concern. ^^ In the context used in this report, ADR 
serves to strengthen and extend the techniques already available 
to the contracting officer for settlement. Given the time-consum- 
ing steps involved in litigation, settlement of the issues between 
the part.ies through use of ADR techniques at least warrants 
thoughtful consideration in most cases. Adding the real cost of 
the litigation risk itself to the comparison of ADR with litiga- 
tion would further tip the balance toward ADR as the less costly 
alternative. 

4 . Short-Term Schedule Problems . 

This objection declares that ADR should be dismissed from 
consideration simply because busy executives cannot spare the time 
required to participate directly in the ADR process. In a sense. 



97 



98 



It is important to hold in mind that the Government rarely 
decides to settle a case simply because it would be expensive 
to litigate the case. There are two key reasons for this. 
First, the Government may not spend funds unless it is auth- 
orized to do so. Second, in the public interest, citizens 
must know that they cannot obtain Government money simply by 
filing suit. Statement of Stuart E. Schiffer, Department of 
Justice, before the Senate Administrative Law and Government 
Relations Subcommittee, Committee on the Judiciary, June 16, 
1988. 

Use of ADR in circumstances other than Government contract 
disputes has resulted in substantial dollar savings over 
litigation. "ADR Is For Real", BNA's ADR Report, Vol. 2, No. 
19, Sept. 1988, at 318. 



184 RICHARD J. BEDNAR 



some forms of ADR does ask busy executives to take on work that in 
the traditional manner of contracts disputes resolution could be 
passed on to litigators. Accordingly, there is a misperception 
that litigation is less disruptive to managers than ADR. 

The answer to this concern is that the busy executives should 
welcome an opportunity to deal with and dispose of a matter early 
through ADR. ADR means early addressal of the disputed issues. 
Litigation on the other hand promises to drain precious, corporate 
and government resources for weeks, months or even years to come. 
Bad news does not improve with age. The more important the 
dispute, the more important it is for the busy executive to become 
involved early in the process. 

The total time and energy required to take on a tough problem 
is likely to be less when concentrated soon after the events, as 
opposed to continued but less concentrated addressal over a longer 
time period. Settlement through ADR suggests prompt and concen- 
trated effort; litigation preparation and execution tends to be 
largely a start-stop effort over a protracted period, with peri- 
odic interruptions of the executive, culminating in many situa- 
tions with a surge effort at the end. In short, whether settled 
by ADR or litigated, a significant contract dispute will put time 
demands upon the executive. 

5. ADR Mav Hide Mistakes From Public Scrutiny . 

It is a legitimate concern that important Government deci- 
sions be shown under the bright light of public scrutiny. Under 
the CDA process this is attained through published decisions of 
boards and courts. Settlement agreements, however, are not 
published. Some argue that settlement of contract disputes are 
essentially secret agreements, hidden from public view and thus 
sheltered from public account£Q>ility. Unlike court or board 
decisions which are published, settlement agreements, so the 
argument goes, may be used to conceal mistakes and bad judgment. 
Others suggest that settling a case makes it easier for the 
Government to give away too much Jn the interest of expediency. 
These concerns can present a substantial obstacle to greater use 
of ADR techniques in support of contract dispute settlement. 

The removal of this obstacle lies in the detail which may be 
required to be included in the written ADR settlement agreement, 
and in what happens to the agreement after it is reached. The 
predicate for public accountability may be established by 
requiring that the basis and justification for the settlement 
agreement be expressed with reasons in written format. Whatever 
the ADR technique, it is important to prepare a memorandum 
reflecting that the settlement result is contractually rational as 
well as fair and reasonable. The memorandum should take the 
reader from the initial position of each contesting participant 
through the negotiating or bargaining process and ultimately to 



ADR IN CONTRACT DISPUTES 185 



the final agreement. In doing this, it is important that the 
memorandum show clearly the reasoning process for arriving at the 
settlement result. The memorandum could then provide written evi- 
dence that the result is rational, fair and reasonable. The evi- 
dence must be detailed enough to reflect the most significant con- 
siderations which shape the settlement agreement, without 
burdening the ADR process with detail or minutia. 

The fulfillment of public accountability may be met by assur- 
ing that the settlement agreement is subject to Government audit. 
Essentially, there is no ultimate difference between a' settlement 
agreement reached through the ADR process and a settlement agree- 
ment reached through traditional CO decision under the CDA or set- 
tlement of quantum issues following litigation of the entitlement 
issues. In all these situations, the ultimate measure of the 
rightness of the result is the competency and integrity of the 
participants . 

6 . The Facts Are Too Complex . 

A critical element of the contract disputes process is know- 
ing the facts . The argvunent which holds up this obstacle to ADR 
asserts that it takes a board or court to thoroughly sort out the 
complex tangle of facts which often characterize a Government con- 
tract dispute. The thought is that ADR necessarily would provide 
a more shallow or too hasty addressal of the facts and thus be 
more likely to lead to a wrong result than that produced by a 
court or board. Just the opposite should be true. 

While a judge unquestionably will know the law, a judge with 
no technical experience in the disputed matters is far less quali- 
fied to deal with the technical issues than are the principals to 
an ADR procedure who have been specially selected for their 
particular knowledge or ability. In other circumstances, the ADR 
principals literally may have lived with the issues as they were 
emerging. Experience with minitrials has demonstrated that even 
principals with thin preknowledge of the case can quickly edssorb 
the ^'acts espential fcr a rational decision. Another important: 
ADR attribute is that no information need be excluded from the 
decisionmakers on procedural grounds, as is true in formal 
litigation. ADR methods facilitate the marshalling and presenting 
of the facts to the decisionmakers. In short, factual complexity 
of the dispute is an argximent favoring ADR rather than favoring 
litigation. 

7. The Law Is Too Complex . 

There are at least a couple of aspects to this objection to 
greater use of ADR. One aspect is that, particularly in view of 
the recent rush of new Government contract laws and regulations. 



186 RICHARD J. BEDNAR 



the legal guidepost.s may be too complicated or undefined to 
support a settlement of the issues. Hence, so goes the argument, 
only the courts and boards should determine what the law is. The 
other aspect of this objection is that settlement is inappropriate 
in those circumstances where legal issues are present as opposed 
to factual issues only. The first aspect of this conceim presents 
the argximent that only courts or boards should establish legal 
precedent . The second aspect presents the notion that legal 
questions themselves are inappropriate for resolution through use 
of ADR. 

The approach to removal of this obstacle is to clarify that 
what is proposed in this report is nothing more than the applica- 
tion of additional, more effective techniques to the settlement 
process which already exists . Government contract controversies 
are settled every day. The CDA, which is the key source of 
procedural law, has been part of the scene for roughly 10 years. 
Most of the CDA is stable law. ADR merely adds sensible refine- 
ment to the settlement process already authorized by the CDA and 
practiced in a more limited way every day. As to the proposition 
that legal precedent is to be made exclusively by the boards and 
the courts, there is no disagreement in this paper. ^^ That, of 
course, means that not all contract disputes are appropriate for 
settlement. If no legal precedent exists, one or both of the 
parties may not be comfortable with settlement. Gaining the 
precedent of a board or court decision may be important and ADR 
would not be pursued in these circumstances. 

On the other hand, as to the argument that ADR techniques are 
inappropriate for deciding legal issues, the answer is that legal 
issues routinely are decided by lay persons, with or without legal 
advice, every day in the ordinary course of Government contracts 
administration. If this were not true, the very process of con- 
tracts administration would come to a standstill while legal is- 
sues such as "latent" defect, "acceptance" of the goods or 
services, "allowable" cost and "final" payment were referred for 
judicial decision. Lay persons can and do decide these legal 
issues every day. Moreover, settling a contract dispute by means 
of ADR does not mean the settlement is accomplished without 
benefit of legal advice. Tne principal representatives of the 
parties to a contract dispute do seek legal advice on legal 
issues. There is every reason why this practice should continue 
under ADR. In short, the fact that legal issues are involved in 
the dispute does not disqualify the case for use of ADR 
techniques . 

In sum, there is a clear, rational path to overcome each of 
these commonly perceived obstacles to wider use of ADR. ADR is 
not here proposed as a panacea to all the frustrations which at- 
tend the CDA disputes process. What is proposed is a logical 



99 

That precedent, of course, can be modified or superceded by 

legislative action. 



ADR IN CONTRACT DISPUTES 187 



extension of existing settlement practice by employment of tech- 
niques which are demonstrably effective and which are gaining 
favor in many other sectors . 



188 RICHARD J. BEDNAR 



CONDITIONS FOSTERING SUCCESSFUL USE OF ADR 

Just as certain preexisting conditions tend to promote suc- 
cessful formation of a Government contract, there are certain con- 
ditions which tend to promote the chances of ADR success in Gov- 
ernment contract disputes resolution. As is true with eveiry con- 
sensual undertaking, the disposition of the participants and the 
climate in which they interact strongly influence the eventual 
outcome. Parties to a Government contract dispute will be moti- 
vated to pursue a particular course of action aimed at dispute 
resolution if they believe that that particular course of action 
is legitimately available and has a reasonable chance of success. 
The parties would not set out deliberately to fail. Other 
factors, of course, bear on selection of a course of action. Time 
and cost are two of these. But the critical element is the belief 
that success is reasonably within reach. These, then, are some of 
the conditions which tend to foster rather than to smother chances 
for ADR success. 

1. Authority . 

The disputes resolution process should be direct and disposi- 
tive. The representatives of the Government and of the contractor 
should have unmistakable authority to reach a binding settlement 
of the disputed issues . They should not be mere spokespersons for 
the real decisionmakers. Disputes resolution discussions between 
persons with authority only to provide advice or recommendations 
are likely to be seen as an unpromising diversion from the 
straighter path to actual disposition by a court or agency board 
of contract appeals. 

The problem with designating persons lacking settlement 
authority to conduct or participate in the ADR proceedings is that 
the results cannot be dispository. At best, such settlement ap- 
proach results in an affirmative recommendation for acceptance and 
approval of a tentative settlement reached by the re.sp«ctive 
authorized participants At worst, it is a fmiitleos exercise 
with no agreement by anybody. Government contracting officers in 
particular tend to staunchly guard their personal authority to 
make "final" decisions in the disputes process. They would tend 
to see an ADR proceeding conducted exclusively by others as either 
an unhelpful exercise or as a threat to their own authority. 

In some ADR circximstances , of course, there would be an 
important role for third-party advisers. In a mini-trial, for 
example, the participation of a qualified neutral adviser can make 
the difference between settlement and failure. Similarly, the 
facts attending the dispute may well make it appropriate for each 



ADR IN CONTRACT DISPUTES 189 



side to gain personal, technical advice during the ADR 
proceeding. ^^^ 

In the ADR process, then, an official empowered to bind the 
Government should represent the Government. This means either the 
contracting officer responsible for the contract out of which the 
dispute arose or an official especially appointed to settle the 
dispute. Similarly, the contractor should be represented by 
souieone with authority to bind the contractor. Authority is seen 
as an important element fostering success in ADR. 

2. Motivation And Commitment . 

The chances of success in using ADR techniques are substan- 
tially improved if the principal representative for each side 
believes that ADR holds a real promise of dispute resolution and 
is worth pursuing. ^^^ This condition is particularly difficult to 
develop in circumstances where the participants have not had ear- 
lier, positive experience with ADR. Because ADR techniques gener- 
ally are not well understood within the Government contracts com- 
munity, there is a natural reluctance to give ADR a try. If the 
disputants don't know about ADR, or believe ADR is not much more 
than an interesting experiment, or think it is a side track to the 
more promising path to dispute resolution, then the ADR approach 
is doomed to fail. There must be sincere motivation to become 
involved in the ADR approach and to see it through to conclusion. 

The motivation to try an ADR approach can be developed. The 
strongest proponents of ADR use in contract disputes are those who 
actually have employed its techniques in contract dispute resolu- 
tion. This speaks well for the future of ADR in this milieu, for 
as greater use of ADR catches on the chorus of supportive voices 
will multiply. Meanwhile, it will remain important for those who 
have practiced or studied the application of ADR to Government 
contracts dispute resolution to continue to explain its virtues 
and to advocate its use. 



100 



101 



But in all these situations, it remains essential that the 
principal representative of each side have actual authority 
to dispose of the dispute. 

In this regard, the earlier in the controversy the better. 
As observed by the Army's Litigation Chief in a letter to 
ACUS dated November 16, 1987, it is difficult to get the par- 
ties to agree to ADR after the contractor has rejected less 
formal procedures by filing suit in the Claims Court. 



190 RICH.\RD J BEDNAR 



3. Approval . 

Conditions must be such that the principal representative of 
each side feels comfortaible that their respective superiors ap- 
prove of the use of ADR techniques. This a separate concern from 
that of authority. Both the contractor and Government par- 
ticipants must be certain that ADR is more than an authorized path 
to dispute resolution; it must also be recognized as an acceptable 
process. The participants seek assurance that they will not be 
criticized for its use. ADR simply cannot work if the principal 
representatives are unsure of their authority or are strongly con- 
cerned that their pxixsuit of dispute resolution through ADR is not 
"okay". In this regard, there is no federal statute or executive 
order or Ck)vemment-wide regulation which specifically sanctions 
and encourages the use of ADR in Government contract dispute reso- 
lution. ^02 

In the absence of such statutory and regulatory underpinning, 
the parties to a contract dispute are likely to lack confidence 
that ADR is "alright", and therefore lack the confidence to try 
it. As a pure legal matter, enabling legislation is unnecessary 
for the practice of ADR (other than, p>erhaps , binding arbitration) 
in the context discussed in this paper. Nevertheless, as is 
discussed more fully in another section of this report, a support- 
ing statute is desirable to place ADR at the same level of ap- 
proval as the primary disputes process prescribed by the CDA. 
This is not to recommend a check list approach to ADR; rather it 
is to suggest authorization and encouragement. An executive order 
implemented by affirmative language in FAR Part 33, "Protests, 
Disputes, and Appeals" also is important to establish a sufficient 
condition of approval and encoxiragement of broader use of ADR 
techniques to resolve Government contract disputes . 

4. ADR Skill . 

Another condition that tends to promote successful use of ADR 
is basic trairing or experience in the skills of .\DP techniques. 
Principal representatives familiar with the m3chanics and art of 
ADR are better equipped than is the novice to embark on a particu- 
lar ADR technique and to move steadily and confidently to dispute 
resolution. Skills such as conflict management and negotiation 
techniques put the parties in a more favorable position to take on 
ADR. Of course, every ADR approach to aispute resolution is a 
unique experience, necessarily shaped and formed by the facts, the 
law and the total make-up of the participants in the particular 
dispute. It is important for the parties to be flexible. Rigid 
protocols are inimical to ADR success. It would be a mistake to 



This thought is developed in a letter from Gary L. Hopkins, 
Associate Counsel, E-Systems, to the author, July 22, 1988. 



ADR IN CONTRACT DISPUTES 191 



develop a stylized routine to be followed in all situations. Nev- 
ertheless, just as there are certain fundamental advocacy skills 
which tend to make one a better advocate, there are certain ADR 
techniques to be learned which tend to make one better in the ADR 
arena. Training in ADR techniques should teach these fundamentals 
in case settings drawn from real situations. At the same time, 
ADR training should emphasize the imperative of flexibility. 

5 . Maneuver Room . 

Contract disputes are unlikely to settle unless there is a 
reasonable range within which each party is willing to come to 
terms. If either party is convinced it has a clear winner, ADR is 
unlikely to work. From an objective viewpoint, plainly meritori- 
ous claims should be paid; conversely, claims clearly without 
merit should not be paid and will not gain in dignity simply 
because they are addressed pursuant to ADR techniques . Both par- 
ties must believe that the correct result is somewhere in the mid- 
dle of a given range of possibilities. ADR works best in an arena 
where claimants with reasonable positions and a willingness to 
make reasonaible compromise are present. 

6. Economy . 

Another important condition which tends to foster use of ADR 
techniques is that the ADR alternative be less costly than litiga- 
tion in either an administrative or judicial forum. This is not 
to suggest that the ultimate compensation paid pursuant to ADR 
must be either greater or lesser than what might be won or lost in 
court. The point here is that the actual cost of the ADR dispute 
resolution process itself is less expensive than the primary forms 
of litigation before a court or appeals board. This is a very 
important factor, for as things are now, all a contracting officer 
need do is make a "final" decision denying the claim, thus leaving 
the litigation effort for others. To the CO, this can be a cheap 
if irresponsible disposition. Managers must be made feunlliar with 
ADR methods, as managers are required by their positions to care 
about the costs of contract disputes resolution in terms of 
resources. For the contracting officer to be willing to find the 
time and make the effort to employ ADR, the contracting officer 
must know that the ADR route in fact will be less costly in time 
and effort than would the full route of litigation. If ADR is 
merely an added step in the process, or if it is perceived as just 
as costly and onerous as litigation, the contracting officer will 
not participate. As those who have had actual ADR experience are 
convinced of its relative cost advantage, the condition of economy 
should not be difficult to demonstrate for parties otherwise 
willing to try ADR. 



192 RICHARD J. BEDNAR 



7 . Contractual Basis For Settlement . 

Another important condition is that the subject matter in 
dispute be resolvable under the Contract Disputes Act of 1978 
(CDA) . Broadly speaking, that means the claims for relief must 
arise under or relate to a Government contract. Additionally, the 
dispute must be one within the subject matter settlement authority 
of the contracting officer under the CDA. This would exclude from 
the ADR process, for examples, claims not arising under or related 
to contract, claims involving fraud and claims involving statutory 
penalties or forfeitures under the jurisdiction of another 
particular federal agency. The essential theme of this report is 
to suggest expanded use of ADR techniques in the course of CDA 
disputes resolution by Government contracting officers; not to 
suggest a widening of the sorts of disputes appropriate for 
settlement by contracting officers. 

8. Settled Government Policy . 

ADR techniques normally would not provide a proper method for 
establishing Government procurement policy. The task to be accom- 
plished is the resolution of a contract dispute, not the estab- 
lishment of new public procurement policy. Those policy matters 
should be left to the traditional agency mechanisms. This is not 
to suggest that ADR is appropriate only in circumstances where the 
law of the case is clearly settled. That is not the condition 
which exists today in many circumstances where a contracting offi- 
cer settles a contract dispute with the contractor. Many contract 
disputes are a tangled mixture of fact and law not surgically sep- 
arcible. Contracting officer decisions do not typically differen- 
tiate between law and fact.^^-^ Nevertheless, contracting officers 
are empowered to settle such cases. Adding ADR techniques to the 
settlement process should not detract from the existing authority 
of contracting officers. 

9. Rules Of Engagement . 

The parties to the contract dispute should agree in advance 
on the frame%^ork within which the ADR procedure will be conducted. 



In other circumstances, contracting officers make legal deci- 
sions directly, such as determining that a given condition is 
a "latent" defect. 



ADR IN CONTRACT DISPUTES 193 



This agreement is best reduced to writing and signed by the prin- 
cipal representative of each party. ^04 while the primary rule 
should be flexibility, a number of procedural and administrative 
issues should be settled in advance in order to facilitate the 
entire process. Among those considerations would be the identi- 
fication of the participants, written expression of the issues to 
be resolved, the date, time and place of the ADR sessions, the 
sequence of events and the manner of payment of the ADR costs . In 
some cases it also may be necessary to reach agreement on the 
extent to which information disclosed during the ADR procedures 
may be used in future related and unrelated proceedings . 



104 

An excellent sample agreement concerning procedures for mini- 
trial appears as an attachment to Turnquist, "ADR initiations 
at the Department of the Navy", Continuing Legal Costs; ADR 
Strategies for Corporations. Law Firms, and Government . 
Edited by: Erika S. Fine, Butterworth Legal Publishers, 1988 
at 305. 



194 RICHARD J. BEDNAR 



COS CAN BE MOVED TO RESOLVE MORE 
DISPUTES WITHOUT RESORT TO LITIGATION 

Generally speaking, the typical CO will make strong efforts 
at resolving the disputed issues before making the final decision 
turning the parties to litigation. ^05 qqs have an enviable record 
of resolving most disputes that come before them. In some 
situations the attempt at settlement will be through exchange of 
correspondence; in other less common situations, the CO and other 
Government representatives will meet with the contractor 
representatives for face-to-face discussions. These are the 
traditional approaches. Rarely does a CO consider alternative 
approaches to dispute resolution. The fundamental reason for this 
mindset, of course, is that traditionally ADR has not been used. 
Before the CDA, the primary disputes process was prescribed by 
regulation and implemented by a contract clause. There was no 
mention of ADR in the regulations or in the contract clauses which 
prescribed the administrative disputes process . Now that there is 
a statutory basis for contract disputes resolution, the CDA, there 
still is no statutory or regulatory language directly authorizing 
or approving ADR techniques for the CO. As a practical matter, it 
simply is too much to expect for most contracting officers 
regularly to initiate an alternate approach to disputes resolution 
when no law or regulation covering disputes even directly mentions 
the subject. 106 

Those agencies which have experimented in any significant way 
with ADR have done so mainly because of the strong influence of a 
high-ranking individual within the agency who has formed an inter- 
est in the subject. ^07 

One way to correct this situation and to encourage expanded 
consideration is through legislation. The Administrative 
Conference has recommended that Congress encourage appropriate use 
of ADR and to make agency disputes resolutions practice simpler 
and more productive. A broad-based bill introduced in the 100th 
Congress by Senator Grassley would confirm agencies' authority to 
employ these methods and encourage them to make use of it. 
S. 2274, the Administrative Dispute ResoJution Act of 1988 . would 



Behre , Arbitration; A Permissible or Desirable Method for 
Resolving Disputes Involving Federal Acquisition and 
Assistance Contracts? (1988), ADR Sourcebook, at 371. 

For ADR to take hold in the contract disputes resolution pro- 
cess it needs to be "institutionalized. " Letter to ACUS from 
Peter H. Kaskell, Senior Vice President, Center for Public 
Resources, November 17, 1987. 

For example, the acknowledged leader of the Government 
contracts ADR effort in the Corps of Engineers is the Chief 
Counsel, Lester Edleman. 



ADR IN CONTRACT DISPUTES 195 



go far to shape a positive government policy encouraging informed 
public sector use of ADR — much as the U.S. Arbitration Act of 
1925 did for arbitration in many private sector disputes. 
Congressman Pease has introduced a companion House ADR bill, H.R. 
5101, that is similar. Even though the bulk of agency comments 
were quite positive, the bills failed to become law during the 
100th Congress. 

The main procedural thrust of S. 2274 and H.R. 5101 would be 
to foster the use of flexible alternatives by allowing the 
contracting parties to shape procedures to fit the circumstances 
on a case by case basis. Although the Grassley proposal would 
potentially apply to all kinds of disputes relating to agency 
administrative programs, it is based wholly on the principle of 
consent — every process addressed in the bill would be invoked 
only with the knowing agreement of the parties, including the Gov- 
ernment . 

The federal Government is in an ideal position to serve as a 
model for the rest of our society in this particular regard. 
Given the CDA policy encouraging settlement. Government contract- 
ing agencies should be in the forefront in using consensual 
dispute resolution. Instead, almost all are lagging far behind 
the remainder of our country's state, local and private sector 
dispute processors. More than one knowledgeable observer has 
noted that while top echelon officials at DOJ sing the praises of 
ADR, all too many litigators cling tenaciously to the motions and 
discovery practice with which they are comfortable and disdain the 
less adversarial approaches. ^^^ This duality should be ended; ADR 
should cease being "one of those subjects that receives almost 
universal endorsement in theory but substantially less in prac- 
tice. "^^^ This attitude should be replaced, with govenuoent 
managers, COs, attorneys, program officials and others viewing 
alternative means of dispute resolution as a major set of tools 
that can be routinely considered and aptly used. 

The impetus for meaningful change will have to come from the 
top levels of government — Congress and top executive officials. 
Otherwise, those who participate in contract disputes resolution 
may prefer the safety of the status quo to thoughtful considera- 
tion of more suitable alternatives. Congress should send a clear 
signal that "ADR is okay", and that informed, rational use of it 
will be supported. S. 2274 is an important first step along this 
path. 



108 



109 



E.g., Richard Mays , ADR & Environmental Enforcement; A Noble 
Experiment or a Lost Cause ?. 18 Envtl. L. Rptr. 10087, 10091 
(1988); remarks of Eldon Crowell at Administrative Conference 
Colloquium on Improving Dispute Resolution, transcript 
(June 1, 1987) . 

Marguerite Millhauser, The Unspoken Resistance to ADR, Nego- 
tiation J. 29 (Jan. 1987). 



196 RICHARD J. BEDNAR 



As directly related to the subject of this report. Section 3 
of the Bill calls on agencies to consider potential ADR uses and 
to develop a dispute resolution policy, to appoint a dispute reso- 
lution specialist, and to ensure personnel training in negotiation 
and dispute resolution methods. These provisions are intended as 
a start to carrying out both the Act's policy favoring ADR and its 
findings that greater use of ADR will improve the operation of the 
government . 

The proposed legislation also cunends the Administrative 
Procedure Act to authorize the parties specifically to agree to 
use mediation, simplified or expedited procedures, or other 
mutually agreeable processes to resolve disputes arising under 
federal administrative programs. (Section 4). Although the use 
of most ADR is not necessarily inconsistent with existing 
requirements, this feature will resolve any doubt about the 
compatibility of ADR with current requirements . Arbitration is 
included subject to general guidelines on issues likely to be apt 
or inapt for ADR (Section 4). 

Section 5 of the Bill amends the Contract Disputes Act to 
encourage agency COs and BCAs to use consensual methods to settle 
accjuisition disputes; it specifically authorizes use of ADR in 
contract disputes, subject to Section 4's guidelines. These chan- 
ges will greatly extend the flexibility of COs, boards of contract 
appeals, and contractors to use minitrials and other appropriate 
means to better handle contract claims. The Act also takes steps 
to make it easier for agencies to use ADR. It authorizes them, 
for instance, to accept volunteer services from mediators or other 
"neutrals" and to hire them promptly and efficiently. These steps 
will encourage other agencies to follow the initiatives of the 
Army Corps of Engineers, the Department of the Navy, and the 
Claims Court in an area where litigation has increased almost 
exponentially in recent years. 

Of course, there are factors other than the absence of a spe- 
cific ADR statute which inhibit the CO's use of alternate appro- 
aches. These were discussed earlier in this report and include 
prominently the specter of being second-guessed or criticized. 
There also will remain even with an ADR statute the comfort of 
following the CDA process knowing that somebody else has to liti- 
gate if the claimant simply is turned down. There also is the 
press on the CO of other business, which leaves little time for 
settling contract claims. Most significant, however, is the fact 
that the primary method of disputes resolution is clearly laid out 
in the statute, the FAR and the contract Disputes clause, while 
ADR is not even mentioned as a possibility. Unless ADR has a 
statutory foundation compatible with the CDA, many COs are 
unlikely to give ADR techniques serious consideration in the con- 
tract disputes process. What this strongly suggests is that ADR 
can never flower in the patch of contract disputes until it is 
planted firmly by statute or at least by some other strong 
expression of federal policy such as an executive order. 



ADR IN CONTRACT DISPUTES 197 



This perceived need for an authorizing statute is not to sug- 
gest that there presently exists a fundamental legal impediment to 
using ADR techniques. With the single possible exception of 
binding arbitration, it does not appear from review of the litera- 
ture and from interviews of experts that there is a statutory 
impediment in the CDA to CO use of ADR techniques to resolve 
Government contract disputes. While legislation may not be needed 
to provide a technical legal undergirding, it could provide a 
clear, unambiguous policy signal that ADR techniques plainly are 
authorized and approved. COs and others need this assurance. 
Until this uncertainly that ADR is alright is removed by legisla- 
tion or executive order, and agencies then encourage COs to employ 
ADR techniques in settling disputes, use of ADR in this context is 
likely to remain at the experimental level. Additionally, as has 
been recommended by the Administrative Conference, ^^^ Congress 
should specifically authorize the use of arbitration in contract 
disputes at the agency's discretion. 

More than Congressional and Presidential anointment of ADR 
will be necessary to realize a significant increase in alternate 
techniques for disputes resolution. The full potential of ADR 
cannot be attained without a cultural change within most contract- 
ing agencies. A clear federal ADR policy and rational implement- 
ing regulations may dissolve agency and CO concern over authority 
for and approval of ADR techniques, but will not have the immedi- 
ate effect of establishing confidence in alternate methods or of 
igniting excitement in their use. There are several practical 
reasons for this. First, it will remain the principal business of 
COs to award new contracts . This concentration on contract 
awards, which cannot be put off, makes disputes resolution, which 
often can be put off, a secondary matter. Traditionally, the job 
performance of COs is evaluated principally on their success in 
contract formation — often rated on the basis of contracts 
awarded or dollars obligated by contract — and for less on their 
achievements in contract dispute resolution. This condition is 
hardly conducive to the trying of more imaginative approaches to 
dispute resolution. This suggests that COs must be specifically 
motivated to use ADR before there can be a dramatic increase in 
its use. 

Effective use of ADR means coming to grips firmly with the 
dispute and reaching a dispositive solution. This is a different 
process than "deciding" a dispute by turning down a claimant and 
forcing litigation. Thus, even with a statute, the COs must be 



110 



Administrative Conference Recommendation 87-5, Assuring the 
Fairness and Acceptability of Federal Agency Arbitration . 1 
C.F.R. S 305.87-5. 



198 RICHARD J. BEDNAR 



further encouraged to try AOR, trained in ADR techniques and 
rewarded for appropriate use of them. 

Some COs too will continue to have the problem of finding the 
time to pursue ADR personally. 

What this all means is that with or without a statute 
providing a firm undergirding, it will take time for the CDA 
mindset to be tempered with the advantages presented by ADR, and 
time to train those involved in the dispute resolution process. 
While ADR training readily can be developed and presented, the 
cultural wariness of ADR will take longer to abate. To mold 
genuine, enthusiastic support for ADR techniques will require 
specific and sustained efforts on many fronts. Both broad 
categories of participants in the contract disputes process must 
be stimulated to reflect upon the possibilities of ADR in each 
particular dispute — Government personnel and contractor 
personnel. Within each of these groupings, the key participants 
tend to mirror each other in job assignment and orientation. 
These key fig\ires are executive officials responsible for the 
contracting process, program and project managers, attorneys, 
accountants, engineers and other technical personnel, and, 
finally, contracting officials. The flow of encouragement should 
be from top-down. Some specific incentives would include: 

1. Stronger federal policy statements — arrived at through 
the FAR process, so as to gain specifically the peirticipation of 
Government contractors and the public — which would encourage 
COs, before final decision, to explore the use of ADR to resolve 
the dispute. 

2. More comprehensive programs of promoting ADR, such as 
that recently established by the Corps of Engineers, to apply ADR 
techniques in specific test cases, to conduct training programs 
and to provide notices and guidance for personnel and 
contractors . ^^2 



The Federal Acquisition Institute (FAI) can be a valuable 
resource in identifying the managerial and operational 
population for whom ADR training is key, and in structuring 
the content of the training programs. Interview of Jack 
Livingston, Director of FAI, Office of Federal Procurement 
Policy, April 29, 1988. 

112 

The Corps of Engineers Program is described in 2 ADR Report 

365 (1988). 



I 



ADR IN CONTRACT DISPUTES 199 



3. Agencies, and perhaps large Government contractors, 
should designate an official within the acquisition hierarchy as 
the AOR Specialist, with the specific mission of developing more 
effective contract disputes resolution practices, to include 
encouraging wider use of ADR methods in the contract dispute pro- 
cess. ^^* 

4. Government contracts should include a clause 
specifically describing the full range of ADR techniques available 
for consideration by the parties at the time a claim is presented 
to the CO for resolution under the CDA.^^^ 

5 . COS involved in the disputes process should be 
specifically evaluated as part of the annual performance 
evaluation cycle on their effectiveness in managing contract 
disputes, to include the use of alternate techniques in disposing 
of such disputes during the rated period. 

6. BCA judges should be specifically alert for suggesting 
return to the CO level of cases which evidently should be pursued 
more vigorously for settlement. ^^^ 



113 



114 



115 



116 



It was reported in The National Law Journal that most 
corporations that seriously pursue ADR designate a "point 
person" who is responsible for conceiving, coordinating and 
implementing the progreun throughout the corporation. 
(Oct. 3, 1988 at 17) 

The ADR Specialist could be a senior acquisition executive 
within the agency. The mission of this position would 
include challenging barriers to wider ADR use and promoting 
ADR use through outreach programs with industry and Govern- 
ment. The fiDR Specialist has potential to achieve substan- 
tially improved understanding of and use of ADR procedures at 
the CO level. The ADR Specialist would operate much as the 
agency "Competition Advocate" operates to foster greater 
competition in contracting pursuant to the Office of Federal 
Procurement Policy Act, 41 U.S.C. S 404. 

The best time to first open the issue of ADR is in the 
drafting of the contract. Reichardt, David L., "Auditing 
Litigation — The Ultimate Finesse, " Contract Management , 
pp. 8-9, July 1986. 

A number of agency boards of contract appeals, to include 
NASA, Navy, Agriculture and Transportation have developed or 
are developing DR rules which could be amended to encourage 
remand to COs in appropriate cases . 



200 RICHARD J. BEDNAR 



7 . ADR training programs for both industry and 
Government personnel will be necessary, as well. In this regard, 
AOR training should not be offered as a "stand alone' subject, as 
though it had some life of its o%^n. Rather, ADR training should 
be integrated into existing training programs on techniques to 
improve the contract formation and contract administration 
processes itself, so as to abate conditions which later lead to 
disputes. ADR training also fits naturally into training on 
broader disputes resolution procedures under CDA. 

In the long run, however, ADR will catch on only to the 
extent that its actual practice produces convincing measvirable 
results . When the parties to a contract dispute come to recognize 
that ADR holds greater promise than does litigation for a just, 
relatively expeditious and relatively less costly solution, ADR 
will become a more popular option in suitable cases. 

There also is the obvious requirement for contractor partici- 
pation in ADR. Will it be difficult to gain the cooperation of 
Government contractors in expanded use of ADR techniques? The key 
is whether contractors can be persuaded as to the practical 
advantages of ADR in given circumstances . ^^^ Government 
contractors are accustomed to conforming with federal policy and 
practices, to include Government procedural requirements. 
Furthermore, the terms of most Government contracts today are set 
by the Government. Pursuant to both the Changes clause and the 
Disputes clause the contractor today does not have the option of 
walking off the job and challenging the Government's position 
through litigation. The Government contract requires that the 
contractor continue performance under the contract, leaving until 
later the determination of the issue of which party will bear the 
costs. ^^5 If contractors can foresee earlier and less costly 
claims recovery through ADR, then ADR will be supported. 



117 



118 



119 



An attachment to the report includes examples of ADR training 
components for COs, recently developed by Wallace Warfield, 
Visiting Fellow, ACUS. 

On the other hand, contractors may be unwilling to agree to 
ADR in circumstances where they have a strong case or where 
they fear "losing face" with the customer by being unwilling 
to compromise. Letter from Robert Eastburn, Jr., Chief Coun- 
sel, Forth Worth Division, General Dynaunics, to the author, 
June 30, 1988. 

Contractors may be expected to be more willing to agree to 
ADR when it is a contractor claim and the Government holds 
the money. Department of Defense Inspector General 
Memorandum for Assistant Inspector General (Analysis and 
Followup) , subject "Review of Contract Audit Reports in 
Litigation", April 1, 1988. 






ADR IN CONTRACT DISPUTES 201 



Contractors tend to be frustrated with the burdens and 
bureaucracy which attend litigation. To the extent that ADR holds 
real promise of fairly resolving the issues in dispute short of 
expensive litigation, the Government contractor should be willing 
to part ic ipate .120 

Other factors also should motivate contractors to participate 
in greater use of ADR. Highly important is that many ADR tech- 
niques keep the outcome within the control of the contracting par- 
ties instead of surrendering the outcome to the board or court. 
Settlement also speeds the process of determining which party 
bears the costs of the contested contract issues, tends to pre- 
serve a cooperative, business relationship between the parties, 
avoids the adversarial nature of litigation, ^21 and fosters the 
early return to concentrating on the real purpose of the contract, 
namely productive contract work. Once ADR is specifically 
sanctioned and encouraged as a preferred choice over litigation, 
willing contractor p£urticipation will be forthcoming. ^22 

CONCLUSION 

The promise of a quick and inexpensive contract disputes 
resolution process that was embodied in the CDA has been 
frustrated by the longstanding practice of overlooking greater 
settlement opportunities at the CO level. Disputes which should 
be settled needlessly pass through the hands of the CO and into 
litigation, feeding the conditions of increased cost and increased 
delay. Litigation costs too much. It takes too long. It tends 
to separate rather than to harmonize the relationships between the 
contracting partners . Greater use of ADR at the CO level holds 
real promise of restoring the original purposes of the CDA. 



120 



121 



122 



Contractor enthusiasm for ADR may be slow-growing. For 
excunple, few Government contractors have agreed to 
participate in the Navy's pilot program of a summary binding 
procedure for small claims at the ASBCA. BNA: Daily Report 
for Executives . Oct. 4, 1988, p. 176. This program, of 
course, is available only after final decision by the CO and 
thus is distinguishable from the ADR use recommended in this 
report . 

Litigation of contract disputes sometimes produces an unnec- 
essarily adversary relationship between opposing counsel. 
vonBaur, F. Trowbridge, The Breakdown of the Changes and Dis- 
putes Process . Pub. Cont. L.J. , 143. 

The disputes clause included in Government contracts is 
intended to provide a quick and efficient administrative 
remedy and to avoid "vexatious and expensive and to the 
contractor often times ruinous litigation." S&E Contractors, 
Inc. V. United States, 406 U.S. 1, 8 ( 1972) (Blackmun, J., 
concurring) . 



202 RICHARD J. BEDNAR 



Greater use of ADR at the CO level can be promoted by 
establishing clear policy signals £rom the pinnacles of the 
Executive and Legislative Branches Government that AOR is okay. 
Practice of ADR methods at the CO level will grow as the 
contracting parties are stimulated and encouraged to employ these 
proven disputes resolution techniques . 



ADR IN CONTRACT DISPUTES 



203 



Contractor 



Government 



REQUESTS FOR PAYMENT 

OR 
CONTRACT ADJUSTMENT 




DEBT COLLECTION 

OR CONTRACT 
ADJUSTMENT ACTION 



CONTROt^ERSY 

OR 

DELAY 



CONTROUERSY OR DELAY 
l»H»r contractor h«s 
opportunity to txprtss 
vievs) 



CLAJM ASSERTED 
(c»rtific*tiof» rrouirrd 
for contractor claims 

evfr $ 50,000) 




CONTBACTOB CLAIMS 

For cltint •{ ySO.OOO 
or undcc ibc CO thcil 
itfuc • dttitimo vilbio 
(0 dtfi. For certified 
claim* ever ^10,000 the 
CO (btll i(*uc I 

drriaioD vilbin (0 dtj* 
at D*ti(f tbc 

ceottictor of tbc 
*rc*iootblc* time 

vilbis wbicb t dccitien 
will be it tued. 




cnaent Contracts Proqrac 
9* Mashlnqton Onlvvrtity 
19S4 



US SUPREME COURT 
(ctrtiorarO 



Difffrent procf dures *pp'y 
for >nn»ss»r Vall*^ Authority 
*nd m»ritim» contract* , 



Ml 



FINAL REPORT ' 

CONFLICT-OF-INTEREST REQUIREMENTS 

FOR MEMBERS OF 

FEDERAL ADVISORY COMMITTEES 



Richard K. Berg 

Senior Counsel, Multinational Legal Services, P.C. 
Washington, D.C. 



May 1989 



This is a report for the consideration of the Administrative Conference of 

the United States. The views expressed are those of the author, and do not 

necessarily reflect those of the Conference, its committees, or staff. 



I 



CONFLICT-OF-INTEREST 207 



Conflict-of-interest Requirements 
for Meabers of Federal Advisory Co— ittees 

The purpose of this study is to examine the conflict-of-interest 
requirements applicable to members of Federal advisory committees. Its scope 
includes not only an analysis of the legal requirements applicable to the 
various categories of members, but an examination of the practices of agencies 
in administering and applying these requirements, with a view to evaluating the 
efficiency and rationality of the present system, and formulating 
recommendations for improvements. 

As the subsequent discussion will demonstrate, there are significant 
semantic problems involved in any discussion of the applicability of conflict- 
of-interest requirements to members of "advisory committees." This paper 
adopts a functional approach and includes within its scope all individuals and 
groups, other than full-time Federal employees, which perform advisory rather 
than operational functions for the Executive Branch or for any Federal agency. 
Even this definition has problems around the edges, but we should be able to 
refine our distinctions as we proceed. 

This paper proceeds from two hypotheses: The first is that the law and 
practices regarding conflict-of-interest requirements have developed from the 
interaction of three statutory schemes, the Federal Advisory Committee Act, 
the conflict-of-interest laws, and the Federal personnel laws, none of which 
was drafted to deal specifically with conflict-of-interest standards for 
advisors to the Government, and that as a consequence this body of law contains 
anomalies and irrationalities which arise from applying classifications and 
distinctions which, however rational in their original context, have little or 
nothing to do with the question of what standards ought to be applied to such 
advisors. The second is that advisory committees perform a variety of roles in 
their relation to the federal decisionmaking process and that the conflict-of- 
interest standard to be applied to any committee member or members should take 
into account the nature of the committee's role and the ability of the 
decisionmaker to discount or offset possible bias deriving from conflict of 
interest. 

Statuiory Frame»fo rk 
Federal Advisory Co— ittee Act 

The Federal Advisory Committee Act (FACA), 5 U.S.C. App. §§1-15, was 
enacted in 1972. It seeks to regulate the establishment, operation and use of 
advisory committees to the Federal Government. The Act's definition of 
"advisory committee" is very broad. It has four key elements. The Act covers 
(1) any committee or similar group, (2) whose members include one or more 
individuals who are not full-time Federal employees, (3) established or 
utilized by the President or a Federal agency, (4) in the interest of obtaining 
advice or recommendations. 



208 RICHARD K. BERG 



The regulatory scheme of the Act covers various aspects of advisory 
committee operations. The Act is aimed at a variety of real and perceived 
problems with Government advisory committees, that some did too much and others 
too little, that there was too little Federal control of their activities, 
review of their effectiveness, or follow-up on their recommendations, that 
advisory committees were not fairly balanced in their membership (particularly, 
in that many were perceived to have an industry bias), and that there was too 
little public access to their deliberations or to the results of their work.-i/ 
To deal with these problems, the Act provided a range of corrective measures. 
The Act requires chartering and periodic review of committees, it attempts to 
ensure greater responsibility in each agency for the operation of the 
committees which the agency establishes, it requires greater openness for 
advisory committees, through provision for publicly noticed and open meetings 
and publicly available minutes. Conflict-of-interest problems were not a 
principal focus of the Act. However, there was very definitely a suspicion 
that advisory committees, particularly in areas involving Federal regulatory 
programs, had an undue industry bias, and, consequently, one of the key 
provisions of the Act was to require that authorities establishing advisory 
committees should "require the membership of the advisory committee to be 
fairly balanced in terms of the points of view represented and the functions to 
be performed," and "assure that the advice and recommendations of the advisory 
committee will not be inappropriately influenced by the appointing authority or 
by any special interest , but will instead be the result of the advisory 
committee's independent judgment." [emphasis added]. 5 U.S.C. App. §5(b),(c). 

The coverage of the Federal Advisory Committee Act is very broad. While 
the term "advisory committee" was familiar in a general way, the classification 
did not previously carry particular legal consequences, and there was no body 
of law to which one could advert to determine what was or was not an advisory 
committee. The Act was designed to regulate agency practices and eliminate 
agency abuses in the establishment and use of advisory committees, and, 
therefore, Congress was unwilling to leave it to the agencies themselves to say 
what they would treat as a covered committee. Consequently, the statutory 
definitions were designed to prevent evasion, and from the start they created 
confusion and controversy. 

The principal areas of confusion over coverage were (1) the so-called 
ad hoc committees or group?, without formal organization, structure, or 
continuing existence, brought together to discuss problems with or present 
views to agency officials; and (2) privately established groups whose views 
were "utilized" by agency officials in the course of their decisionmaking. 

Initial confusion over the status of ad hoc groups and "utilized" 
committees was increased by early judicial rulings giving a broad inter- 



im See, e.g. . S. Rept. No. 92-1098, 92d Cong., 2d Sess. 3-4, Federal Advisory 
Committee Act Sourcebook, 155-56. 



CONFLICT-OF-INTEREST 209 



pretation to the statutory definition and by the failure of the agencies 
charged with providing guidance under the Act to do so.^ 

Finally, in response to a 1980 recommendation of the Administrative 
Conference of the U.S.,^ the General Services Administration in its 1983 
interim guidelines and its 1987 final rule on Federal Advisory Committee 
Management^ attempted to narrow the coverage of the definition. To deal with 
the problem of the ad hoc groups, the GSA regulation excludes meetings with 
Government officials "for the purpose of exchanging facts or information," 
meetings initiated by the group for the purpose of presenting its views to an 
official or agency where the official does not use the group "recurrently as a 
preferred source of advice," and meetings initiated by a Federal official with 
more than one individual, where the official is seeking individual and not 
consensus advice or recommendations .-2/ To deal with the problem of "utilized" 
committees, GSA adopted the view of the ACUS recommendation that a privately 
established committee is not "utilized" within the meaning of the Act unless 
the agency through institutional arrangements "adopts [the committee] as a 
preferred source of advice." 41 CFR §§101-6.1003, 6.1004. 

The definitions in the GSA guidelines have provoked little controversy and 
have been useful in excluding from the scope of the Act the ordinary informal 
give-and-take between Government decisionmakers and the private sector. Yet 
they assuredly do not eliminate all problems of coverage. Indeed, they 
introduce new bases of distinction which in operation are likely to prove 
somewhat nebulous. What exactly is a "preferred source of advice?" Is there a 
real distinction between "obtaining the advice of individual attendees" and 
seeking "to obtain consensus advice or recommendations?"^ 

Underlying these largely semantic questions is a basic policy question. 
Is FACA to be viewed as limited to those advisory committees which might be 
termed "official" or "quasi-official" in that they are organized by or receive 
financial and/or logistical support from the Government? Or is FACA to be 
viewed as a more ambitious undertaking, as basically a requirement governing 



2/ See National Nutritional Foods Assn. v. Califano . 603 F.2d 327, 334-36 
(2d Cir. 1979), reviewing the legislative, administrative and judicial 
authorities. 

y Recommendation 80-3, 1 CFR §305.80-3. 

1/ 41 CFR Part 101-6, 52 F.R. 45926. 

5/ 41 CFR §101-6. 1004(i), (j), (1). 

^1 See NRDC v. Herrington, 637 F.Supp. 116 (D.D.C. 1986) holding that members 
of a scientific panel brought together to advise the Department of Energy in an 
examination of the safety of DOE's Hanford, Wash, plutonium reactor were not an 
advisory committee because they were being asked to provide individual views 
and not to act as a committee. The Court indicated that it was not disposed to 
interpret FACA broadly in situations which do not raise the problems that 
Congress was concerned about when it enacted FACA. 



210 RICHARD K. BERG 



substantially all situations in which the Government actively seeks advice from 
the private sector? Some aspects of FACA point in each direction. The 
provisions on committee charters, periodic review, etc., make sense only as 
applied to those committees which are, realistically, a part of Government. 
The requirements with respect to openness and membership balance seem to speak 
more broadly to the entire process by which Government receives outside advice. 
So long as this dualism remains, questions of coverage are likely to persist. 

One noteworthy fact about FACA and its administration is that the Act 
purports to apoly equally to all covered advisory committees regardless of 
their function. 2/ Obviously, advisory committees are established and employed 
for many different purposes: to tap a source of expertise not available within 
the Government, to secure the views of interest groups affected by Government 
policy, to provide an objective critique of a Government program or activity, 
to develop a broad consensus on a delicate or controversial issue, and even to 
provide an excuse for inaction. 

Various attempts have been made to break the committees down into rough 
functional categories. Prof. Cardozo listed five functional categories: 
policy advice, technical advice, fact-finding, evaluation of proposals and 
applications, and advice in personnel selection. 2/ Wegman cites four 
categories: peer or grant review committees, technical or program review 
committees, special fact-finding or investigative committees, and general 
policy advisory committees.^ The Annual Report of the President on Federal 
Advisory Committees for Fiscal Year 1987, prepared by GSA's Committee 
Management Secretariat, lists six functional groups for committees: 
scientific/technical program, non-scientific program, grSOt review, major 
national policy/issue, regulatory negotiation, and "other". 10/ 

While these categorizations are helpful in describing what it is advisory 
committees do, they cannot weigh precisely the relative significance of 
committees' recommendations to the decisionmaking process. It seems likely, 
however, that the most influential committees, i.e. those whose recommendations 
correlate most closely with the final governmental action, would be the grant 



•Z-/ However, judicial decisions suggest that only parts of the Act may be 
applicable to "utilized" committees, ?ee. Cent er for Auto Safety v. Cox . 
580 F.2d 689, 604 (D.C. Cir. 1978); Washington Legal Foundation v. American Bar 
Association . 648 F.Supp. 1353, 1360-61 (D.D.C. 1986). 

^' Cardozo, The Federal Advisory Committee Act in Operation . 33 Ad. L. Rev. 
(1981), 1980 ACUS 313, 343. 

2./ Wegman, The Utilization and Management of Federal Advisory Committees 
(1983) pp. 29-31. If we lump Prof. Cardozo's last two categories, which are in 
some respects similar, his classifications and Wegman's are essentially the 
same. 

iO' Sixteenth Annual Report of the President on Federal Advisory Committees, 
p. 2. The breakdown is based on the sponsoring agency's categorization. The 
regulatory negotiation category accounted for less than 1% of the total. 



CONFLICT-OF-INTEREST 211 



review committees and those committees which provide a technical expertise 
which the Government does not have in-house. Conversely, on matters of broad 
policy the governmental decisionmaking authority, whether an agency, the 
President or Congress, is less likely to defer to the committee judgment, and 
the committee's influence is likely to depend on the persuasiveness of its 
report, the political acceptability of its proposals, and a variety of other 
considerations. Obviously, the greater the committee's influence on the 
decisionmaking process, the more important it becomes to prevent conflicts of 
interest. 

For purposes of this paper, the point to be emphasized is that FACA, and 
particularly FACA in its broader reading, embraces committees and similar 
groups with a great variety of functions, organizational structures and ties, 
relationships and obligations to the Federal Government. Indeed, perhaps the 
only element common to all covered advisory committees is that in some way and 
for some reason they are looked to by the Federal decisionmaker as "a preferred 
source of advice. " 

Conflict-of-interest Requirements 

FACA, as we have seen, prescribes no conflict-of-interest requirements for 
advisory committee members. Indeed, it might be argued that it adopts an 
alternative approach for obtaining objective advice. FACA's requirement that 
committees be "fairly balanced in terms of the points of view represented," is 
tacit recognition that members are not expected to approach their agenda with 
Olympian detachment, although the very next statutory injunction, "that the 
advice and recommendations * * * will not be inappropriately influenced by the 
appointing authority or by any special interest, but will instead be the result 
of the advisory committee's independent judgment," certainly asks the members 
to walk a very thin line. 

At any rate, FACA was not enacted in a vacuum, and it may be assumed that 
the drafters were aware of the general framework of legal requirements 
governing conflicts of interest of advisory committee members. The most 
significant of those requirements are those imposed by 18 U.S. Code Chapter 11, 
entitled "Bribery, Graft, and Conflicts of Interest." 

18 U.S.C. Ch. 11 is applicable to Federal employees generally, aiid many of 
its provisions go back to the nineteenth century. However, the statutes were 
thoroughly revised in 1962 by Public Law 87-849. One of the principal 
contributions of P.L. 87-849 was to distinguish for purposes of conflict-of- 
interest requirements between full-time Federal employees and "those who serve 
the Government intermittently or for a short period of time," and to except 
persons in the latter category from certain of the prohibitions imposed on 
ordinary full-time employees. 

To this end P.L. 87-849 created the category of "special Government 
employee" (hereinafter SGE), defined as "an officer or employee * * * who is 
retained, designated, appointed or employed to perform, with or without 
compensation, * * * temporary duties either on a full-time or intermittent 
basis" for not to exceed 130 days in any 365-day period, 18 U.S.C. §202(a). 



212 RICHARD K. BERG 

The various prohibitions of 18 U.S.C. Ch. 11 are complicated, but may be 
described generally as followsril/ 

18 U.S.C. §201 prohibits bribery of public officials and witnesses in 
Federal proceedings. It punishes both the giving and the receiving of bribes. 
The term "public official" includes any person "acting for or on behalf of the 
United States * * * in any official function." The term has been interpreted 
broadly, see Dixson v. United States . 455 U.S. 482 (1984), il/ and probably 
would include members of Federal advisory correnittees who are SGE's and arguably 
even those who are independent contractors. It is unlikely that this provision 
would cover one who is advising as a representative of an outside group. 

18 U.S.C. §§203 and 205 deal with representation of a private party by a 
Government employee in a judicial or administrative matter in which the United 
States is a party or has a direct or substantial interest. A regular 
Government employee may not represent another person before an agency or a 
court in such a proceeding, with or without compensation. An SGE is subject to 
much more limited prohibitions. (1) He may not represent anyone else before a 
court or agency in a matter in which the United States is a party or has a 
direct and substantial interest where the matter involves a specific party or 
parties and where he has at any time participated personally and substantially 
in the same matter in the course of his Government employment. (2) If the SGE 
has served in his agency more than 60 days in the past 365 days he may not 
represent anyone in a matter before that agency involving a specific party or 
parties. (It should be noted that in conflict-of-interest parlance, particular 
matter involving a specific party or parties is a relatively narrow concept and 
does not include a rulemaking of general appl icability. )ii/ 

18 U.S.C. §207 deals primarily with post -employment restrictions. With 
respect to both regular and special Government employees, section 207(a) 
imposes a permanent ban on representation in connection with a particular 
matter involving a specific party or parties, where the employee participated 
personally and substantially in the decisional process in the course of his 
Government service. Section 207(b) imposes a two-year ban on representation 
with respect to a particular matter involving a specific party or parties when 
the matter was actually pending under the former employee's official 
responsibility within one year prior to his leaving Government. It also 
imposes a two-year ban on certain kinds of assistance in representation by so- 



il/ Chapter 11 consists of sections 201-219 and 224. However, only sections 
201-03. 205, 207-09, 218 and 219 have relevance to the subject of this paper. 

i^' In Dixson the Court held, 5-4, that officers of a private non-profit 
corporation administering community development grants as the designee of a 
local governmental grantee were "public officials" under section 201. 

13./ There is considerable overlap between sections 203 and 205. According to 
the explanatory memorandum of the Department of Justice, "for all practical 
purposes Section 205 completely overshadows Section 203 in respect of officers 
and employees of the Government." Section 203. however, also covers Members of 
Congress. 



CONFLICT-OF-INTEREST 213 



called Senior Employeesii/ with respect to particular matters involving 
specific parties in which they participated personally and substantially while 
in Government employment. Section 207(b) does not exclude SGE's from its 
operation, but it is unlikely to impact very frequently on SGE's because they 
are rarely Senior Employees and do not usually have such particular matters 
pending under their official responsibility. ("Official responsibility" is 
defined, §202(b), as "direct administrative or operating authority" and 
presumably would not include the responsibilities of an employee serving in an 
advisory capacity.) 

Section 207(c) applies exclusively to Senior Employees, including SGE's 
who serve for more than 50 days in a calendar year, and forbids for one year 
after employment with the agency has ceased any representational activity 
before that agency irrespective of the subject matter. 

Section 207(g) is the only provision in section 207 not addressed to post- 
employment activities. It prohibits a partner of a Government employee from 
acting as agent or attorney before any agency or court with respect to any 
particular matter (not qualified by "involving a specific party or parties") in 
which the employee has participated personally and substantially. This 
provision is specifically applicable to SGE's. Because the "particular 
matters" to which it applies need not involve specific parties, section 207(g) 
has the anomalous effect of prohibiting representation by a partner of an SGE 
in circumstances in which the SGE himself would not be barred by section 203(b) 
or section 205(b). 

18 U.S.C. §208 is the most important conflict-of-interest provision for 
purposes of our inquiry. Section 208 prohibits a Government employee, 
including an SGE, from participating "personally and substantially" as a 
Government employee "through decision, * * * recommendation, the rendering of 
advice, investigation, or otherwise, in * * * [any] particular matter in which 
to his knowledge, he, his spouse, minor child, partner, organization * * * has 
a financial interest." This is an especially broad prohibition in several 
respects. "Particular matter" is not qualified by the words "involving a 
specific party or parties" and has been interpreted very broadly by the 
Department of Justice. i^ It does include adoption of rules of general 
applicability. Second, the participation which is forbidden includes the 
rendering of advice. Third, the ranqe of financial interests it reaches is 
broad, and there is within the statute itself no de nn'nimis exception. 
However, because of its very breadth the statute itself recognizes a need for 



ii' "Senior Employee" is a term defined in 0PM regulations, 5 CFR §737.25, to 
refer to those employees specified in or designated pursuant to 18 U.S.C. 
§207(d)(l). 

i^' In a 1978 opinion the Office of Legal Counsel, Department of Justice 
stated, "[W]e have consistently interpreted §208(a) to apply to rule-making 
proceedings or advisory committee deliberations of general applicability where 
the outcome may have a 'direct and predictable effect' on a firm with which the 
Government employee is affiliated, even though all other firms similarly 
situated will be affected in a like manner," 2 Ops. OLC 151, 155 (1978). 



214 RICHARD K. BERG 



exceptions. Accordingly, section 208(b) provides for waivers by the employing 
agency (1) in particular cases where the employee discloses to the appointing 
official the conflicting financial interest and the appointing official makes a 
written determination that "the interest is not so substantial as to be deemed 
likely to affect the integrity of the services" of the employee, and (2) where 
the agency by general rule exempts certain financial tnterests as "too remote 
or too inconsequential to affect the integrity" of employees' services. is/ 

18 U.S.C. §209, which forbids a Government employee to receive any salary 
or supplement as compensation for his services to the Government, is 
specifically made inapplicable to SGE's. 

If we consider the potential impact of 18 U.S.C. §§201-09 on SGE's who are 
advisory committee members, it is evident that the principal trouble spot is 
section 208. Section 201 is broad in coverage, but hardly a trap for the 
unwary. The prohibitions in sections 203 and 205 are narrow, and, at least, as 
applied to lawyers, no more than common sense and the Canons of Ethics would 
require. i^ The revolving door provision of section 207(a) is likewise narrow 
and reasonable, and the broader prohibitions of sections 207(b) and (c) are not 



^ Where agencies have used their authority to exempt by general rule under 
section 208(b)(2) the exemption has almost invariably been for described types 
of financial holdings, such as bank deposits, insurance policies, Government 
and municipal bonds, and interests in diversified mutual funds. See, e.g. . 
15 CFR §0.735-13 (Dept. of Commerce); 40 CFR §3. 301(b) (EPA); 24 CFR §0.735- 
204(d) (HUD); 16 CFR §5. 8(d) (FTC). The Administrative Conference of the 
United States, however, has adopted a section 208(b)(2) rule defining certain 
agency actions as too remote from a member's financial interests to affect the 
integrity of his services. 1 CFR §302. 5(b)(2). This was done after the 
Department of Justice had ruled that public members of the Conference were 
special Government employees and that section 208 might reach a member's action 
in voting on a recommendation for procedural changes in an agency before which 
he practiced. 

■i^ Sections 203 and 205 do apply to any kind of representation, not merely 
representation as an attorney. For example, one who negotiates a contract with 
the Government on behalf of a corporation acts as an agent within the meaning 
of section 205. However, section 205 states that neither that section nor 
section 203 prevents an SGE from acting as agent or attorney for another in the 
performance of work under a Government grant or contract where the head of the 
department or agency certifies that the national interest so requires. But 
this exemptive authority may have been significantly limited by the 1988 
amendment to the Office of Federal Procurement Policy Act, P.L. 100-679, §6, 
41 U.S.C. 423(e), which prohibits any Government employee who participated 
personally and substantially in the conduct of an agency procurement from 
participating personally and substantially in the performance of the contract 
arising out of such procurement. 



CONFLICT-OF-INTEREST 215 



likely to apply to many advisory committee members. ia/ Section 207(g), the 
prohibition on representational activity by a partner of an employee, is worth 
special attention because it may come into play in unforeseen and rather 
innocent circumstances, particularly where an advisory committee member is a 
member of a large law firm. However, section 207(g) applies primarily to 
lawyers in private practice, and a relatively small number of committee members 
are lawyers. i^ 

Section 208, on the other hand, is very broad, as I have pointed out. 
Since it covers rulemaking of general applicability, as well as substantially 
any other kind of agency action, an S6E who serves on an advisory committee 
which is making a recommendation respecting a rule that would affect the 
industry to which he has financial or employment ties would be bound to recuse 
himself unless he received a waiver from the agency under section 208(b). 

The Federal Personnel Manual recognizes this problem and encourages the 
use of waivers where "the special Government employee renders advice of a 
general nature from which no preference or advantage over others might be 
gained by any particular person or organization." FPM, p. 735-C-4. The Office 
of Legal Counsel has observed that this provision of the Manual constitutes a 
"gloss" on the statutory language of section 208(b)(1) that "the interest 'is 



i^ Another post-employment restriction with potential application to some 
advisory committee members is contained in the new procurement integrity 
section of the Office of Federal Procurement Policy Act, 41 U.S.C. §423, added 
by P.L. 100-679, §6. Section 423(e) provides that "no Government official 
* * * who has participated personally and substantially in the conduct of any 
Federal agency procurement" shall for two years thereafter either participate 
in any manner on behalf of a contractor in negotiations leading to the award, 
modification or extension of the contract which is the subject of the 
procurement or participate personally or substantially in the performance of 
such contract. This prohibition applies to special Government employees and 
thus could, in theory at least, reach SGE's who serve on advisory committees. 
Whether an advisory committee would ever be sufficiently involved in the 
procurement process to meet the personal and substantial participation test of 
section 423(e) is a more difficult question. An OFPP guidance memorandum of 
Ap'-il 7, 1989 on section 423 states, "We 'expect that special Government 
employees would rarely participate personally and substantially in the conduct 
of a procurement." However, it seems clear from the statute that participation 
includes acting in an advisory capacity, see §423(n)(3)(B) , so that the 
possibility of the prohibitions applying to members of an advisory committee 
cannot be ruled out. There is no provision in section 423 for waiver of its 
substantive prohibitions. (Section 423 was scheduled to take effect May 16, 
1989, but legislation has just been enacted extending the effective date 60 
days to enable the Administration to explain its terms to covered employees and 
contractors. ) 

i^./ According to the Seventeenth Annual Report of the President on Federal 

Advisory Committees, only 763 (3.8%) of the approximately 20,000 members of 

advisory committees are lawyers. Presumably this figure includes Government 
lawyers. 



216 RICHARD K. BERG 



not so substantial as to be deemed likely to affect the integrity of the 
services which the Government may expect' from the special Government employee" 
and seems particularly geared to "members of advisory committees, who are often 
specifically chosen because of an expertise that results from this affiliation 
with particular organizations, firms, or groups having a general interest in 
the very matters before the advisory committee." However, OLC emphasized that 
the exemption process "compels the responsible Agency official to focus on the 
questions of the special Government employee's outside affiliations and to make 
a specific written finding with respect to the expected integrity of the 
individual's services. "-^^ 

Two other provisions of 18 U.S.C. Ch. 11 are worth brief mention. 
Section 219 makes it a criminal offense for a public official to act as an 
agent of a foreign principal required to register under the Foreign Agents 
Registration Act of 1938, 22 U.S.C. §§611-621. The prohibition applies to 
special Government employees, but section 219(b) authorizes a waiver for an SGE 
where "the head of the employing agency certifies that such employment is 
required in the national interest." Although the Foreign Agents Registration 
Act does not apply to ordinary legal representation before courts and agencies, 
it does apply where the representation includes "attempts to influence or 
persuade agency personnel or officials other than in the course of established 
agency proceedings," 22 U.S.C. §613(g); see 28 CFR §5.306. Approximately 100 
law firms are registered under the Act. The Act also reaches legislative 
lobbying and public relations activities on behalf of foreign principals. 

Finally, 18 U.S.C. §218 authorizes the President and the heads of Federal 
departments and agencies to "declare void and rescind any contract, loan, 
grant, subsidy, license, right, permit, * * * ruling, decision, opinion or rate 
schedule * * * in relation to which there has been a final conviction for any 
violation of this chapter * * *." However, the possibility for voiding a 
Government decision tainted by a violation of the conflict-of-interest laws is 
not limited to the situation in which there has been a criminal conviction. 
Prior to the enactment of P.L. 87-849, the 1962 legislation, the Supreme Court 
had held in the well-known Dixon-Yates case. United States v. Mississippi 
Valley Generating Co. . 364 U.S. 520 (1961), that the United States was entitled 
to cancel a contract in the negotiation of which there had been a violation of 
the predecessor statute to section 208, notwithstanding that no criminal 
prosecution had ever been brought. In its explanatory memorandum on 
P.L. 87-849, the Department of Justice stated with respect to section 218 that 
since the powers it grants are "in addition to any other remedies provided by 
law" it would not seem to override the Dixon-Yates case. At least one case 
has held that the Government may disavow a contract where there was a violation 
of section 208 notwithstanding the absence of a criminal conviction, K. & R. 
Engineering Co. v. United States . 616 F.2d 469, 472-75 (Ct. Claims 1980); see. 



^/ 2 Ops. OLC 151, 156-57 (1978). If the Manual provision in question is 
read as exclusively applicable to section 208(b)(1), it would follow that 
waivers in such circumstances could be granted ad hoc but not by general rule. 
The Nuclear Regulatory Commission has stated that it is its policy to grant 
waivers to SGE's in the circumstances described, but such waivers must 
nevertheless be granted individually. 10 CFR §0.735-21(d) . 



CONFLICT-OF-INTEREST 217 



also, United States v. Medico Industries . 784 F.2d 840, 844-46 (7th Cir. 
1986). 

While the potential for the Government's voiding its prior action or 
decision on the basis of a violation of the conflict-of-interest laws should 
not be overlooked, the likelihood that a violation by a member of an advisory 
committee would lead to such a result is rather slim, particularly so, where 
the advice rendered is on a broad policy issue, rather than on the disposition 
of a particular matter involving specific parties, such as a contract, grant or 
license. 

Disclosure Requirements . In addition to the criminal prohibitions of 
18 U.S.C. Ch. 11, SGE's are subject to certain financial disclosure provisions. 
Under Executive Order 11222 of May 8, 1965 and 0PM regulations, 5 CFR §735.412, 
SGE's are required to file with the employing agency a confidential statement 
of other employment, including corporate directorships, trusteeships and 
similar affiliations, and such financial information as the appointing agency 
determines is relevant to the SGE's duties.^!/ SGE's who are paid at the rate 
of a GS-16 or higher also must file the more elaborate public financial 
disclosure form (SF-278) required of high level Government officials by Title 
II of the Ethics in Government Act. Since the disclosure requirements are in 
addition to and not in lieu of the prohibitions in 18 U.S.C. Ch. 11, disclosure 
of a disqualifying financial interest does not of itself work an exemption from 
section 208. The burden is on the employee to obtain the appropriate waiver. 

The Office of Government Ethics is at present revising the regulations 
governing the system of confidential financial reporting by Federal employees, 
including SGE's. The proposed regulations published December 2, 1986, would 
have expanded the reporting requirements for SGE's well beyond what has been 
generally required under Executive Order 11222, but because of provisions 
permitting agencies with OGE approval to "tailor" reporting requirements to 
their needs by asking for either more or less information and to exempt some 
individuals entirely, it is hard to gauge the impact precisely. The proposed 



^/ Section 306 of E.O. 11222, which prescribed financial disclosure 
requirements for SGE's, was stricken from the Order by E.O. 12565 of 
September 25, 1986 in order to permit development of a comprehensive system of 
financial reporting for employees in the Executive Branch under the Ethics in 
Government Act. See text at note 22, infra . However, the OPM regulation 
implementing section 306, 5 CFR §735.412, has remained in effect, as have the 
practices of nearly all agencies in requiring SGE's to file confidential 
statements. 



218 RICHARD K. BERG 



regulation has met with considerable criticism, and one may anticipate 
significant changes in the final version. ^^ 

The applicability of the provisions of 18 U.S.C. Ch. 11, as well as the 
financial disclosure requirements cited above, to SGE's depends exclusively on 
the SGE's employment status and not (except for the "responsibility" provision 
in section 207(b)) on the nature of their functions. In other words, these 
conflict-of-interest statutes do not distinguish between operational and 
advisory responsibilities. Nevertheless, Congress was well aware that SGE's 
frequently serve in advisory roles, and indeed the principal purpose of 
creating the category of SGE's was to ease the recruitment of experts and 
consultants whose main work was performed outside of Government. 

Status of Advisory Co—ittee Meabers 

As we have seen, the category of special Government employee was 
established to deal with the conflict-of-interest problems arising from the 
situation of part-time or intermittent Government employees, serving with or 
without compensation, frequently in advisory roles. But because some part-time 
advisors are to be treated as SGE's, it does not follow that all must be. The 
variety of advisory relationships between persons in the private sector and 
decisionmakers in the Federal Government has been the source of some confusion. 

The principal guidance to the agencies on the subject of the status of 
members of Federal advisory committees for purposes of the conflict-of-interest 
laws is a memorandum dated July 9, 1982 from the then Director of the Office of 
Government Ethics, J. Jackson Walter, to the heads of all Executive departments 
and agencies..^' This memorandum reviewed at length the administrative 
practice in dealing with part-time Government advisors since shortly before the 
enactment of P.L. 87-849. The principal distinction which the Walter 
Memorandum pointed to was that between those advisory personnel who are 
selected because of their individual qualifications and those who are expected 
to act in a representative capacity for an industry, for labor or agriculture 
or some other group. This distinction, first enunciated in a memorandum from 
President Kennedy dated February 9, 1962 and entitled "Preventing Conflicts of 
Interest on the Part of Advisers and Consultants to the Government," has been 



i2J The proposed regulation, 51 F.R. 43359-65, would require covered 
employees, including SGE's, to report the source and nature (but, apparently, 
not the amount) of any investment income, and a description (but, apparently, 
not the value) of other assets held for production of income, the source of any 
earned income, and information respecting employment or affiliations with 
profit or non-profit organizations. Agencies would, however, have authority to 
exclude individuals from "all or a portion of the reporting requirements" on 
the basis of appropriate findings by the agency head. The regulations would 
exclude from coverage individuals "performing services for the United States as 
an independent contractor under a personal service contract." 

^' As the memorandum points out, "There is no substantive difference between 
an appointee providing advisory service individually and one doing so as a 
member of a committee." Walter Memorandum, p. 2. 



CONFLICT-OF-INTEREST 2 1 9 



maintained with revisions to take into account the provisions of P.L. 87-849, 
in the regulations of the Office of"*fersonnel Management, Federal Personnel 
Manual, Part 735, Appendix C. The pertinent portion of Appendix C is worth 
quoting at length because, together with the Walter Memorandum, it remains the 
principal operative guidance to the agencies. 

"It is necessary occasionally to distinguish between 
consultants and advisers who are special Government 
employees and persons who are invited to appear at an 
agency in a representative capacity to speak for firms or 
an industry, or for labor or agriculture, or for any other 
recognizable group of persons, including, on occasion, the 
public at large. A consultant or adviser whose advice is 
obtained by an agency from time to time because of his 
individual qualifications and who serves in an independent 
capacity is an officer or employee of the Government. On 
the other hand, one who is requested to appear before a 
Government agency to present the views of a nongovernmental 
organization or group which he represents, or for which he 
is in a position to speak, does not act as a servant of the 
Government and is not its officer or employee. He is 
therefore not subject to the conflict of interest laws and 
is not within the scope of this chapter. 

The following principles are useful in arriving at a 
determination whether a person is acting before an agency 
in a representative capacity: 

(1) A person who receives pay from the Government for 
his services as an adviser or consultant is its employee 
and not a representative of an outside group. The 
Government's payment of travel expenses and a per diem 
allowance, however, does not by itself make the recipient 
an employee. 

(2) It is rare that a consultant or adviser who 
serves alone is acting in a representative capacity. Those 
who have representative roles are for the most part persons 
serving as members of an advisory committee or similar body 
utilized by a Government agency. It does not follow, 
however, that the members of every such body are acting as 
representatives and are therefore outside the range of the 
conflict of interest laws. This result is limited to the 
members of committees utilized to obtain the views of non- 
governmental groups or organizations. 

(3) The fact that a person is appointed by an agency 
to an advisory committee upon the recommendation of an 
outside group or organization tends to support the 
conclusion that he has a representative function. 

(4) Although members of a governmental advisory body 
who are expected to bind outside organizations are no doubt 
serving in a representative capacity, the absence of 
authority to bind outside groups does not require the 
conclusion that the members are Government employees. What 
is important is whether they function as spokesmen for non- 



220 RICHARD K. BERG 



governmental groups or organizations and not whether they 
can formally commit them. 

(5) When an adviser or consultant is in a position to 
act as a spokesman for the United States or a Government 
agency — as, for example, in an international conference--he 
is obviously acting as an officer or employee of the 
Government." 

The Walter Memorandum recognizes, however, that the distinction set out in 
Appendix C between SGE's and private sector representatives does not exhaust 
the possibilities for classifying those who furnish advice to the Government. 
Advisory services may also be furnished through a contract, and in that case 
the advisor or advisors may be independent contractors. 

One of the key indicia of the employment relation in the Government, as 
elsewhere, is that an employee must carry out his duties under the supervision 
of another employee. An independent contractor, the Walter Memorandum points 
out, "is not hired under the civil service laws and is not subject to the 
supervision that inheres in an employee-supervisor relationship in the civil 
service. More to the point, he is not an employee for the purposes of 
18 U.S.C. §§202-09." 

The Memorandum observes that the employee/ independent contractor 
distinction is not often a problem because "committees are rarely brought into 
the service of an agency by means of a contract." But if the problem were to 
come up, the issue would be resolved under the ordinary legal principles which 
govern the distinction between employees and independent contractors. 

"However, it is worthwhile to mention an issue that 
could arise in connection with the conflict-of-interest 
statutes if an agency were to create an advisory committee 
and then enter into a contract with it or each of its 
members individually. The issue is whether the agency 
would in practice exercise supervision over the operations 
of the committee and the formulation of judgments by its 
members that was great enough to taint the contract as a 
device for concealing their true status as SGE's under 
§§202-209. If an agency, for example, were to convene a 
committee and award the members a contract pursuant to 
which they (1) produced, after independent study, an 
advisory paper dealing with a problem that the agency's 
staff was too busy to resolve on its own and (2) delivered 
the paper without antecedent clearance from the staff or 
agency head, the committee members would properly have been 
deemed contractors. However, if the committee worked 
routinely subject to the scrutiny of the staff and with a 
significant amount of guidance from it, the members would 
be open to the charge that they actually served as SGE's 
and were subject to §§202-209. As appears from these 
examples, the question is one of degree." 



CONFLICT-OF-INTEREST 221 



Still another distinction discussed in the Memorandum is the circumstance 
in which a Federal official receives "unsolicited, informal advice from an 
outside individual or group of individuals regarding a particular matter or 
issue of policy that is within his official responsibility. Or he may himself 
bring up an agency matter of policy issue informally with one or more outsiders 
in order to obtain their views. An incident of this sort sometimes prompts the 
inquiry whether the outsiders have become SGE's of the agency. In general, the 
answer is that they have not, for they are not possessed of appointments as 
employees nor do they perform a federal function." However, Mr. Walter felt 
compelled to subject this conclusion to a caveat : 

"An official should not hold informal meetings more or 
less regularly with a non-federal individual or group of 
individuals for the purpose of obtaining information or 
advice for the conduct of his office. If he does so. he 
may invite the argument that willy-nilly he has brought 
them within the range of 18 U.S.C. §§202-09. " [emphasis 
added]. 

This portion of the Memorandum seems a departure from what is otherwise a 

very tightly reasoned discussion. It seems very doubtful that the mere 

frequency of meetings of one or more Federal officials with a group of 

individuals can "willy-nilly"^^ and contrary to the intentions of all 
concerned bring those individuals into an employment relation with the 

Government so as to subject them to the criminal sanctions of 
18 U.S.C. §§202-09. 

What we have here is basically the problem of the ad hoc advisory 
committee. But if we assume that at some point the informal ad hoc committee 
shades over into a committee covered by the FACA, does it necessarily follow 
that at the same point on the spectrum the committee members become SGE's? 
And if at that point the Federal official should comply with FACA and formalize 
these informal arrangements, are these committee members SGE's if he fails to 
do so? We must bear in mind that FACA and the conflict-of-interest laws have 



^/ The term is well chosen for it comes from the Latin volens/nolens . 
will-ing or not willing. The notion of an advicor becoming a Federal employee 
"willy-nilly" has given the Department of Justice pause. "In the usual case 
[the employment] relationship is based on an identifiable act of appointment. 
However, an identifiable act of appointment may not be absolutely essential 
* * * in a particular case where the parties omitted it for the purpose of 
avoiding the application of the conflict-of-interest laws or perhaps where 
there was a firm mutual understanding that a relatively formal relationship 
existed." 1 Ops. OLC 20, 21 (1977). 

In this opinion the Office of Legal Counsel concluded that a personal 
friend of the President who advised him almost daily on an informal basis on "a 
wide range of policy issues" did not thereby become a special Government 
employee. However, when the adviser called and chaired meetings attended by 
employees of various agencies and assumed responsibilities for coordinating 
Administration activities on a particular issue, OLC ruled that the adviser was 
engaged in a Governmental function and should receive a formal appointment. 



222 RICHARD K. BERG 



somewhat different purposes, that they impose different obligations and 
burdens, and that violations of their provisions entail very different 
consequences. 

The Walter Memorandum does not address still another problem raised by 
FACA, the members of "utilized" committees. The reason is doubtless that these 
committee members are assumed not to be subject to the conflict-of-interest 
laws because, by hypothesis, they have not been selected and, therefore, not 
appointed by a Federal official. 

To sum up, in theory there appear to be at least five categories of 
members of Federal advisory committees in terms of conflict-of-interest law 
analysis (excluding regular Government employees serving in their official 
capacities): 

1. Members selected by a Federal official to serve on a formally 
established advisory committee in their individual capacities with or without 
compensation. 

2. Members selected to serve in a representative capacity, as described 
in Part 735, Appendix C of the Federal Personnel Manual. 

3. Independent contractors who have been engaged by the Government to 
supply advice or recommendations pursuant to a contract and whose work is not 
supervised or controlled by a Federal employee. ■22/ 

4. Members of an informal ad hoc group whose advisory activities have 
crossed the FACA threshold but who have not been "appointed in the civil 
service" within the meaning of 5 U.S.C. §2104 and §2105. 



15/ There is some question whether a committee which performs consultant 
services for the Government under a contract is an advisory committee under 
FACA. See Lombardo v. Handler, 397 F.Supp. 792, 797-800 (D.D.C. 1975) aff'd 
without op., 546 F.2d 1043 (D.C. Cir. 1976), cert, denied, 431 U.S. 932 (1977), 
citing legislative history to the effect that FACA "does not apply to persons 
0" organizations which have contractual relationships with the Federal agencies 
nor to advisory committees not airectly established by or for such agencies." 
397 F.Supp. at 799, citing 118 Cong. Rec. 31421. On the other hand, 
Paul Weiss, GSA's Associate Administrator for Administration, recently stated 
in responding to a Congressional question regarding the applicability of FACA 
to advisory committees created in connection with the performance of Federal 
contracts: "[T]he test for inclusion under the Act is that the agency's 
contract requires that the advice of a group or other than contractor personnel 
be provided to the agency, and that the contractor has not merely solicited 
advice as part of the activities incidental to the performance of the 
contract." Hearing before the Senate Committee on Governmental Affairs, 
Department of Defense/Strategic Defense Initiative Organization Compliance with 
Federal Advisory Committee Act (4-19-88). 

Agency responses to Senator Glenn's questionnaire, see text at note 31, 
infra , classified a few committee members as independent contractors. 



CONFLICT-OF-INTEREST 223 



5. Members of committees which are privately established but which are 
"utilized" by the Government as "a preferred source of advice." 

The first category of advisory conmittee members are SGE's and are subject 
to the conflict-of-interest laws. All the other categories of members are not 
SGE's and are not subject to the conflict-of-interest laws. 

This multiplicity of categories is in itself neither good nor bad. Most 
subjects in Government are complicated. The Federal Advisory Committee Act 
covers a very broad range of situations and relationships, and it is not 
surprising that in dealing with conflict-of-interest questions, distinctions 
must be made. 

There are, however, two points which must strike an observer: 

(1) The operative distinctions seem primarily conclusory and almost 
invariably questions of degree. Take, for example, the distinction -- crucial 
for purposes of conflict-of-interest coverage — between a member selected as 
an individual and one selected as a representative. There are undoubtedly 
situations in which an agency is primarily interested in ascertaining the views 
of the industry or other interest group from which the member is selected. 
Some statutes establishing advisory committees even specify the representative 
character of the member ship.-^o/ But agencies can generally obtain industry 
views on an issue without establishing an advisory committee. One would assume 
that in many situations what the agency is looking for is just the kind of 
borderline or hybrid situation the conflict-of-interest laws appear to rule 
out, that is to say, the honest judgment of someone who is forced to weigh his 
perception of the common good against the narrower interests or biases of the 
constituency from which he is selected. ^Z/ Of course, the agency can resolve 
the problem on the legal level simply by classifying the members as 
representatives while hoping that in practice they can rise above mere 
representation. But it does seem to me, at least, that the either/or 
classification called for by Appendix C and the Walter Memorandum over- 
simplifies what is likely to be a spectrum of situations and a mix of motives 
on the part of all the participants. 

One result of the difficulty in classifying advisory committee members is 
that the question of compensation -- the only "hard" indicium -- tends to 
bccomo determinative. An uncompensated adviser may be an SGC, but a 
compensated adviser must be an SGE. Appendix C states flatly that "[aj person 
who receives pay from the Government for his services as an adviser or 



^/ See, note 29, infra . 

■^' One might well ask, if what is desired from the member is a mere 
expression of the industry view, why have him work on a committee? The 
committee device implies an interest in developing a consensus through exchange 
of views, which in turn implies that the individual's own views are of interest 
to the agency. Of course, it may well be that the function of the committee is 
not so much to communicate the constituencies' views to the agency as to make 
the agency's actions more palatable to the constituencies. 



224 RICHARD K. BERG 



consultant is its employee and not a representative .of an outside group." 
Therefore, if an agency desires to compensate a committee member, he must be 
treated as an SGE, even if his role on the committee is to "balance" the views 
of members serving as representatives of special interests. 

While the question of compensation should doubtless be relevant in 
determining whether an adviser should be considered an SGE, I doubt that the 
flat rule stated in Appendix C is desirable or realistic. Presumably the 
agency benefits from advice received from persons acting in a representative 
role or capacity. Why should the agency be forbidden to pay them to do so? In 
other contexts such payments are allowed, as when an agency provides financial 
support for public interest representation in its proceed ings.-^a/ Indeed, when 
Congress has turned its attention to the problem in the course of establishing 
statutory advisory committees, it has on occasion provided for compensation for 
committee members who serve as representatives of outside interests. ^^ After 
all, it is not the fact of payment which creates the conflict of interest, but 
rather the existence of a conflict which is assumed to make payment 
inappropriate. In the case of representatives deliberately chosen as such, 
this assumption seems of doubtful validity. 

Even more nebulous in this context is the distinction between the SGE and 
the independent contractor. If the key issue is supervision and control, the 
requirement of FACA that recommendations of the advisory committees "not be 
inappropriately influenced by the appointing authority * * * but will instead 



^/ Opinion of the Comptroller General B- 139703 (July 24, 1972); see, 
generally, Boyer, Expense-Reimbursing Public Participants in Administrative 
Rulemaking: The Federal Trade Commission Experience, 1979 ACUS 437, 438-46. 

^' The legislation establishing the Technical Pipeline Safety Standards 
Committee, 49 U.S.C. App. §1673, and the Technical Hazardous-Liquid Pipeline 
Safety Standards Committee, 49 U.S.C. App. §2003, specifically provides 
compensation for the non-Federal members of the committees, and that such 
payments "shall not render members of the Committee employees or officials of 
the United States for any purpose." Similar provisions are in the legislation 
establishing the National Boating Safety Advisory Council, 46 U.S.C. §13110, 
and the National Highway Safety Advisory Committee, 23 U.S.C. §404. The 
Department of Transportation treats these members as representatives. The 
statute establishing the Advisory Council on Employee Welfare and Pension 
Benefit Plans, 29 U.S.C. §1142, provides that three members "shall be 
representatives of employee organizations" and three "shall be representatives 
of employees." It also provides that members shall be entitled to daily pay at 
the rate of a GS-18. The Department of Labor has felt compelled to classify 
these members as SGE's notwithstanding their representative role on the 
committee, and has provided appropriate waivers under section 208. On the 
other hand, where section 7 of the Occupational Safety and Health Act, 
29 U.S.C. §656, authorized, but, in Labor's view, did not compel compensation 
for advisory committee members serving as representatives, Labor decided not to 
provide compensation in order to treat the members as representatives and not 
as SGE's. 



CONFLICT-OF-INTEREST 225 



be the result of the advisory committee's independent judgment" seems 
inconsistent with the conclusion that committee members are SGE's. 

(2) But aside from the difficulties of applying these distinctions, there 
is the troubling question whether they make any sense in terms of the problem 
of conflicts of interest.' The distinction between advisory committee members 
who are selected to exercise independent judgment and those who are selected to 
represent distinct interests is sound enough in theory whatever its 
difficulties of application. The problem is that it is not realistic in terms 
of the purposes of the agency in establishing the committee. Distinctions 
between SGE's and independent contractors in terms of conflict-of-interest 
requirements make little sense, even in theory. With respect to members of ad 
hoc or of utilized committees, the distinction can probably be justified by the 
practical impossibility of bringing the conflict-of-interest laws into play 
where the relationships to the Government are so tenuous. But that does not 
eliminate the danger of self-serving advice or obviate the need for some 
alternative mechanisms. 

One of the issues I have researched is how agencies have administered the 
fine distinctions set forth in the Walter Memorandum. Given the number of 
agencies and committees involved, 1&/ it has simply not been feasible to examine 
on any systematic basis the correctness of the legal conclusions the agencies 
have reached. However, the information available suggests a tendency of 
agencies to adopt a policy of administrative convenience and to consider 
members as interest group representatives, at least in all doubtful cases. 

In March, 1988, Chairman Glenn of the Senate Committee on Governmental 
Affairs sent a letter to the agencies asking twenty questions respecting their 
management of advisory committees. One question was: "Do you consider members 
of advisory committees to be regular government employees, special government 
employees, or independent contractors? What criteria has the agency 
established to make this determination?" I have examined the responses to this 
question from 25 agencies and obtained additional information through inter- 
views at a half-dozen agencies. My conclusions are as follows: 

1. Agencies tend to place (or at least, to report) all or nearly all non- 
Government members in the same category. Thus, the Department of Health and 
Human Services, the agency with the most ronmittees and o\'er 20% of the total 
committee members, reported, "Members of committees which are established to 
advise the Department are routinely appointed as special government employees." 
Other agencies which reported that all or nearly all members were SGE's 
included the Department of Education, NASA, the United States Information 



■^' There were 992 advisory committees and 57 sponsoring agencies in fiscal 
1987. Total membership was 19,837 individuals, of whom approximately 8% were 
full-time Federal employees. There is no breakdown on the status of the 
remaining membership. Sixteenth Annual Report of the President on Federal 
Advisory Committees, pp. 1-2, 7, 9. In fiscal 1988, 58 departments and 
agencies sponsored 1,020 committees, and a total of 21,236 individuals served 
as committee members. Seventeenth Annual Report of the President on Federal 
Advisory Committees, p. 1. 



226 RICHARD K. BERG 



Agency, and the Nuclear Regulatory Commission. On the other hand, the 
Department of Transportation, with over 20 committees and over 500 members, 
reported that all its advisory committee members served as representatives, and 
the Department of the Interior, with approximately 150 committees and over 
1,500 members, reported that the members of only one committee were considered 
SGE's; all other committees were made up of members serving in a representative 
capacity. Other agencies reporting that their members were all or mostly 
representatives included the Department of Energy, the Department of the 
Treasury, the Office of the United States Trade Representative, the Eximbank, 
the Federal Home Loan Bank Board, the Federal Mediation and Conciliation 
Service, and the Small Business Admini strati on. 11/ Although many agencies 
cited the 1982 Walter Memorandum as their principal source of guidance, my 
impression (possibly unfair) is that of the agencies whose responses I 
reviewed, only the Department of Labor and EPA have attempted to apply the 
criteria on a case-by-case basis. Four agencies reported that some committee 
members were serving as independent contractors. 

2. In at least two instances, committee members who were clearly SGE's 
under the Walter Memorandum and FPM guidelines were classified as 
representatives for reasons of administrative convenience. (Neither instance 
involved significant potential for conflict of interest.) Several agencies 
reported as a basis for not treating members as SGE's that they received no 
compensation for their services, a consideration which should not be 
controlling under the Walter Memorandum. 

3. Agencies, by and large, do not demand financial disclosure from 
committee members who are not SGE's. The National Endowment for the Humanities 
classifies its advisory committee members as independent contractors. However, 
since the committees make recommendations with respect to award of grants, the 
Endowment reported that it took precautions to avoid appointing panelists to 
review applications from institutions with which they were affiliated. (The 
replies are too sketchy to get an adequate idea of agency efforts to brief non- 
SGE members on conflict problems.) 

Availability of Waivers 

While the distinctions in classifications of advisory committee members 
seem conf-jsing and artit'-ary, the waiver authority in 18 U.S.C. §208 does offer 
an avenue of relief from the threat of criminal liability, and it Is used, at 
least in some agencies. However, there appear to be a number of drawbacks: 

(a) Some agencies, notably the Department of Defense, are extremely 
reluctant to issue section 208 waivers. This policy may increase the pressure 



■^' These agency replies expressly or impliedly exclude from consideration 
committee members who are full-time Federal employees. However, the membership 
figures cited, which I have drawn from the Sixteenth Annual Report, are total 
committee membership. 



CONFLICT-OF-INTEREST 227 



on those officials responsible for establishina advisory committees to find 
ways to avoid classifying their members as SGE's.^2/ 

(b) Doubts have been expressed whether a section 208(b) waiver may be 
granted where the financial interest in question is not insubstantial in an 
absolute sense, even though the agency concludes that any bias arising from 
that interest may be offset through committee balance or, perhaps, discounted 
as a result of disclosure. Because of these doubts the President's Commission 
on Federal Ethics Law Reform is recommending an amendment to section 208 to 
authorize waivers for advisory committee members where the appointing authority 
determines, after review of the financial disclosure forms, that the need for 
the member's expertise outweighs the potential for conflict of interest. The 
Commission concluded that under the present statute, "officials with the 
authority to grant waivers are required to look in large measure at the 
magnitude of the financial interest in question. "-^l/ 

I do not believe the administrative' interpretation of section 208 is as 
narrow as the Commission assumes. In the 1978 Office of Legal Counsel opinion 
previously discussed, -M/ OLC addressed the issue of substantiality of interest 
and concluded on the basis of the contemporaneous understanding of the 1962 
legislation and the subsequent administrative construction in the Federal 
Personnel Manual that a waiver may be granted to an advisory committee member 
even though his financial interest is not insubstantial in an absolute sense 
provided that he "renders advice of a general nature from which no preference 
or advantage ^QXer others might be gained by any particular person or 
organization."-^ Nevertheless, agencies may not be sufficiently aware of 
their authority in these circumstances, or they may feel unwilling to subscribe 
to a determination that the conflicting interest is not "likely to affect the 
integrity" of the advice to be obtained. (In this respect the Commission's 
formulation -- "the need for a member's expertise outweighs the potential for 
conflict of interest" -- seems a more accurate statement of the basis for the 
determination. ) 

(c) The waiver process is not always smooth. In a case brought to the 
attention of the Glenn Committee the President's AIDS Commission had been 
assembled and had conducted several public meetings before it was realized that 
the members' affiliations with health care providers, medical research 



^ The President's Commission on Federal Ethics Law Reform points out that 
"an undesirable stigma already attends the seeking of waivers," and it stated 
its belief that "waivers should be granted more frequently instead of 
discouraged." These comments were directed at the administration of 
section 208(b) generally, but they are also relevant to advisory committees. 
To Serve with Honor . Report of the President's Commission on Federal Ethics Law 
Reform (1989), p. 24. 

il/ Ibid ., p. 30. 

•31' See text at note 20, supra . 

^1 2 Ops. OLC 151, 156 (1978), quoting from the FPM Ch. 735, App. C. 



228 



RICHARD K. BERG 



institutions and other private entities might constitute conflicting financial 
interests. Waivers were hastily granted, but o 
unknowingly, been exposed to criminal liability.- 



interests. Waivers were hastily granted, but onlv after the members had, all 



(d) Waivers tend to be limited in scope and to place the burden on the 
member to anticipate potential conflicts outside that scope. Let me quote from 
a sample waiver letter to a committee member from the Secretary of Labor: 

"I have determined pursuant to 18 U.S.C. 208(b)(1) 
that the personal financial interests you have disclosed 
are not so substantial as to be deemed likely to affect the 
services you will provide * * * when those services are on 
matters of broad policy or general applicability. On these 
matters the provisions of 18 U.S.C. 208(a) are waived. 
This waiver would not extend to matters which may have a 
direct, predictable and unique effect on your financial 
interests or the interests of other related and associated 
persons covered by 18 U.S.C. * * *." 

Should a member's lawyer advise him to take the job under such a limited 
waiver? It is true that the waiver tracks the line of distinction set forth in 
the FPM and the OLC opinion. But it is the agency and not the advisory 
committee members which is supposed to set the committee's agenda, and it seems 
reasonable to expect the agency to be in a better position than the individual 
member -- once financial disclosure has been made -- to anticipate and guard 
against conflicts of interest. Accordingly, where waivers are used, it would 
be far better for the agency to set out in the waiver what the committee is 
expected to do, what conflicting interests have been disclosed, and its 
conclusion that so long as the committee stays within the former, the interests 
disclosed shall not be deemed disqualifying. 

(e) Finally, waivers address only the section 208 problem. This is 
admittedly the most serious one, but there are others, including a minor 
irritant, the requirements of the Hatch Act. which would apply on any day when 
the member is performing committee service. ^2/ 

Another such problem is the potential impact of the new procurement 
irtayrity provision of the Office of Federal Procurement Policy Act.-^s/ This 
legislation contains restrictions Oii employment discussions between 
"procurement officials" and competing contractors during the conduct of a 
procurement, on the unauthorized disclosure of certain information regarding a 



■^' U.S. General Accounting Office, Report to the Chairman, Committee on 
Governmental Affairs, U.S. Senate, GA0/6GD-89-17 AIDS Commission Compliance 
with FACA. 

iZ/ The Hatch Act, 5 U.S.C. §§7324-27, restricts partisan political activity 
by Federal employees. It applies to SGE's only on the days they perform 
Government services, 5 CFR §733. 123(b). 

^ 41 U.S.C. §423. See note 18, supra . 



CONFLICT-OF-INTEREST 229 



procurement, and on the post-employment activities of Federal officials 
involved in a procurement. Whether this legislation will affect any advisory 
committees will depend on how broadly the procurement process is defined, and 
thus far this definition has proved elusive.-^ To the extent the statute 
applies to advisory committees, the applicability of the revolving door 
provision, §423(e), and possibly the restriction on employment discussions, 
§423(a), (b), will depend on whether the committee member is regarded as an 
SGE. Significantly, section 423, unlike 18 U.S.C. §208, contains no agency 
authority to exempt insignificant conflicts of interest. 

Conclusions and Recommendations 

1. The principal conclusion of this study is that the present 
distinctions among advisory conmittee members between special Government 
employees, on the one hand, and representatives and independent contractors, on 
the other, are extremely fine, difficult to administer consistently, and, in 
any event, cannot be justified in terms of protecting the Government from self- 
interested advice. In practice they result in agencies requiring too little 
disclosure from the members who are not SGE's, while imposing significant 
burdens, principally potential criminal liability, on those members who are 
SGE's. 

2. The burdens associated with classification as an SGE, in terms of 
financial disclosure, potential criminal liability. Hatch Act coverage, 
contrast sharply with the minimal burdens on committee members who serve in any 
other capacity. Consequently, at least where committee service is not 
compensated, agencies are likely to avoid classifying members as SGE's. 

One must recognize, of course, that the likelihood of actual criminal 
prosecution for an inadvertent violation of 18 U.S.C. Ch. 11 is remote. It is 
Department of Justice policy to bring criminal charges under these provisions 
only in situations involving aggravating circumstances. Nevertheless the 
possible embarrassment associated with a criminal investigation is something 
most committee members would seek to avoid. Furthermore, the Government has an 
obligation to its advisory committee members, many of whom serve for no or 
merely nominal compensation, not to expose them to even theoretical risks. 
Finally, overbroad and unrealistic prohibitions tend to debase the currency of 
ethics rules and may make the public indifferent to even serious violations. 

3. Advisory committees vary greatly in their functions and in the degree 
to which conflicts of interest must be guarded against. Generally speaking, 
the need to avoid conflicts is greatest with respect to those committees 



^' The drafters of the interim regulations to implement the procurement 
integrity provision confessed that "the vastly different types and nature of 
procurements conducted throughout the Federal Government make it impossible to 
identify one event that is common to all agencies, that constitutes the 
beginning of a procurement," 54 F.R. 20488. Nevertheless, the interim rule 
states that a procurement begins when an authorized agency official "determines 
that a specific agency need or requirement shall be satisfied by procurement," 
48 CFR §3.104-7, 54 F.R. 20494. See, also, section 3.104-4(h)(2 



i 



230 RICHARD K. BERG 



involved in reviewing grants or similar applications and those furnishing 
scientific or technical advice, whereas for those committees providing policy 
advice the avoidance of conflicting personal interests is less important than 
assurance that the agency is aware of those interests and able to discount 
possible bias where appropriate or to offset it through committee balance. 

The thrust of the proposed recommendations is to: 

(a) Require from each member of an advisory committee (excluding full- 
time Government employees) the same level of disclosure of employment, 
affiliations and financial interests, irrespective of the member's status as an 
SGE, contractor, or representative. The agencies should have authority to 
exclude from the disclosure requirement information, other than the 
individual's employment, not deemed relevant to the purposes and functions of 
the committee, much as they do now with respect to SGE's under former 
section 306 of E.O. 11222. Thus, where the committee is dealing with social 
rather than economic issues, or where the impact of its recommendations on 
identifiable financial interests is likely to be remote, it would be 
appropriate to require no financial information. Any willful misstatement in 
the disclosure document would be punishable under 18 U.S.C. §1001. 

(b) Replace the present unsatisfactory bases of distinction among 
advisory committee members with specific authority in the agency to classify 
members of advisory committees as SGE's or non-SGE's on the basis of the 
function of the committee and the agency's expectations of the degree of 
objectivity to be expected of the member. In general, members of those 
advisory committees which advise on the disposition of particular matters 
involving specific parties, e.g. . grant review committees, scientific advisory 
groups,^' would continue to be classified as special Government employees, 
whereas members of those committees which advise on matters of general policy 
would not. But the choice in each instance would be an explicit one by the 
agency, after consultation with the Office of Government Ethics. 

The recommendation is limited to those committees formally established and 
whose members are selected by the Federal Government. It is, therefore, not 
coextensive with the coverage of the Federal Advisory Committee Act, nor would 
it apply to those ad hoc or informal committees which fall outside the coverage 
of FACA. 

(c) The final paragraph is intended to deal with the anomaly in 18 U.S.C. 
Ch. 11 described on page 7 of the report. Because section 207(g) applies to 
"particular matters," while sections 203(c) and 205 apply more narrowly to 
representation with respect to particular matters "involving a specific party 
or parties," a partner of a special Government employee is at present under a 
broader disability with respect to representation than is the employee himself. 

Finally, we can all agree, at a minimum, that if no changes are made in 
the coverage of the conflict-of-interest laws, as proposed herein or by the 



— ' E.g. . the Nuclear Regulatory Commission's Advisory Committee on Reactor 
Safeguards, which is an integral part of the NRC's licensing process, 42 U.S.C. §2039. 



CONFLICT-OF-INTEREST 231 



President's Commission, the agencies must do a better job of advising committee 
members and proposed committee members of their potential exposure. However 
remote the prospect of prosecution, the potential is there, it could lead to 
embarrassment, if nothing worse, and the Government owes it to these members, 
many of whom volunteer their services without compensation, to warn them of the 
risics and how they may be avoided. Agencies should make clear to each advisory 
committee appointee what his legal status is to be, and if he is classified as 
an SGE, he should be briefed on his conflict-of-interest exposure, not merely 
in general terms, as is frequently done, but also with particular reference to 
the anticipated agenda of the committee. One of the purposes of the Federal 
Advisory Committee Act was to increase agency control over the activities of 
its advisory committees, and with that control should go the responsibility for 
guiding the committee members through the pitfalls which their status entails. 



REFORMING ASYLUM ADJUDICATION: ON NAVIGATING 
THE COAST OF BOHEMIA* 

David A. MARTiNf 

* © Copyright 1990. David A. Martin. 

t Professor of Law, University of Virginia. This Article is a revised and 
abbreviated version of a consultant's report that I presented to the Administrative 
Conference of the United States in May 1989. The views expressed here are my own 
and do not necessarily reflect the views of the Conference or of its committees. TTie 
formal recommendation adopted by the Conference on the basis of the report was 
published at 54 Fed. Reg. 28,964, 28,970-72 (1989) (codified at 1 C.F.R § 305.89-4 
(1990)). 1 have expanded my treatment of these issues, and have, incorporated 
considerable additional information on the asylum adjudication systems of several 
European countries and Canada, in a book to be published by Yale University Press. 

The research support, advice and comments provided by the members and staff 
of the Administrative Conference have been of real assisunce and are gratefully 
acknowledged. The other individuals whose help made this work possible, 
particularly government ofiicials, private lawyers, and scholars, in Europe, Canada, 
and the United States, who generously sacrificed large blocks of time for interviews 
for this study or for review of the manuscript, are too numerous to list here, but their 
assistance is deeply appreciated. I am also especially grateful to the German Marshall 
Fund of the United States and the Ford Foundation, whose financial support made 
possible the European research, principally during sabbatical leave from law school 
teaching in 1984-85, when I was in residence at the Henry-Dunant Institute in 



Table of Contents 

I. The Substantive Legal Framework 

A. International Provisions 

B. Anurican Legal Provisions 

II. The Poucy Context 

A. Angles of Vision 

1 . The Asylum Tradition 

2. The Need for Control; Asylum as a Loophole . . 

B. Factual Issues 

1. Lack of Clarity Concerning the Substantive 
Leg^ Standard 

2. The Coast of Bohemia 

a. The Essential Problem 

b. Bmces vs. Spectrums 

c. Lessons 

3. Limited Accessibility of the Facts on Which to 
Base an Adjudication 

a. Adjudicative Facts 

b. Legislative Facts 

c. Predictive Ju/igment 

4. DifRculties of Cross-Cultural Communication . . . 

C. The Imperative of Speedy Final Decisions 

1 . The Scope of the Magnet ElfFect 

2. The Alternative of Deterrent Measures 

3. Toward a Better-Targeted Deterrent 

III. The American Adjudication System 

A. Historical Background 

B. The Current System 

1 . District Office 

2. Immigration Court 

3. State Department Role 

4. Administrative Review 

a. The Board of Immigration Appeals 

b. The Asylum Policy and Review Unit 

5. Judicial Review 

6. Actual Deportations 

Geneva. Special mention must be made of Dr. Jiri Toman, deputy director of the 
Institute, whose unfailingly gracious assistance is well-known to nearly anyone who 
has undertaken humanitarian-law research, and of Rita Trigo Trindade of Geneva, 
my European research assistant, whose multilingual talents and diligent probing 
were indispensable. 



REFORMING ASYLUM ADJUDICATION 235 

7. TheRoleoftheUNHCR 

8. The August 1987 Proposed Regulations 

C. Evaluation 

1 . Speed and Fairness 

a. Two Bites at the Apple 

b. Administrative and Judicial Review 

c. Is Delay Really a Problem? 

d. Rights to Counsel 

2. Accuracy 

a. Diffusion of Responsibility 

b. Political Bias 

c. Inadequate Use of Existing Expertise 

3. Consistency and Quality Control 

IV. Proposed Reforms 

A. Specialized Adjudicators 

B. Organizational Location 

C. Nature of the Proceedings 

D. Fairness and the Treatment of Legislative Facts 

E. Administrative Review 

1 . Administrative Review or Not? 

2. The Recommended Framework 

3. Country Guidelines 

F. Judicial Review 

G. Deportation 

H. Emergency Responses to Large-Scale Influxes 

1 . Quick Denial of Manifesdy Unfounded 
Applications 

2. Disqualification for Transit Through Third 
Countries 

3. Ending Work Authorizations and Making 
Alternative Arrangements for Subsistence 
Pending Adjudication 

V. Conclusion 

Over the last decade, sharply rising numbers of asylum seekers 
have placed g^eat strains on the Western democracies' venerable 
practice of granting political asylum. In part, the new difficulties 
derive from the growing penetration of law into what once had been 
a largely discretionary practice; in part, they result from improved 
global mobility and communications. None of the challenged adju- 
dication and protection systems have responded well, and some have 
responded abysmally. 



236 DAVID A. MARTIN 

Critics abound, offering widely variant views on what are the 
most serious problems besetting these systems. Some find the major 
problem to be that "economic migrants" or "false refugees" are able 
to gain years of delay, no matter how weak their asylum claims, by 
abusing the asylum process's abundant procedural protections, 
adopted to ensure against returning someone to immediate persecu- 
tion. Others find the most serious failings to lie in governments' 
reflexively negative reactions to new arrivals, at least if they are of 
the wrong ideological stripe. Overly restrictive legal doctrine, 
impossibly demanding standards for factual proof, and a proliferat- 
ing array of barriers and deterrents are seen as the major problems,* 

In this country, controversy over political asylum waxed and 
waned throughout the 1980s. The decade opened with bitter con- 
tention, when the 1980 boadift of 125,000 Cubans from the port of 
Mariel coincided with high levels of asylum applications from Hai- 
tians, Nicaraguans, and Iranians. In the succeeding years, the Rea- 
gan Administration implemented new restrictive policies, including 
detention of many asylum applicants and the interdiction on the high 
seas of boats coming from Haiti. It thereby managed, through the 
middle of the decade, to keep application levels relatively modest 
and to relegate asylum to the shadows occupied by low-priority polit- 
ical issues. Refugee advocacy groups nonetheless kept up a volley of 
denunciations aimed at the new restrictions. They charged that 
these policies, and the low acceptance rates of applicants from Cen- 
tral America and the Caribbean, reflected political manipulation or 
racism or simple heartlessness. These groups complained that the 
government's practices violated legal obligations founded in statute 
and treaty.^ In a pattern that has been long familiar in American 
asylum policy, several rounds of relatively unproductive litigation 



• See generally Martin, The New Asylum Seekm, in The New Asylum Seekers: 
Refugee Law in the 1980s 1 (D. Martin cd. 1988) (tracing the history of refugee 
protections from World War II to the present, highlighting new problems, and 
suggesting solutions). 

2 See, e.g.. G. Lx>escher & J. Scanlan, Caix:ulated Kindness: Refugees and 
America's Half-Open Door, 1945 to the Present 170-208 (1986) (recounting 
controversies over asylum policy that brought into question the Carter and Reagan 
Administrations' commitment to the ideological neutrality of the 1980 Refugee Act); 
Meissner, Rejections on the Refugee Act of 1 980, in The New Asylum Seekers: Refugee 
Law in the 1980s 63-64 (D. Martin ed. 1988) (criticizing the failure "to administer a 
judicious political asylum policy" for Salvadoran refugees, as prescribed by the 1980 
Refugee Act); Scanlan, Regulating Refugee Flow: Legal Alternatives and Obligations Under 
the Refugee Act of J 980, 56 Notre Dame Law. 618, 620-23 (1981) (discussing alleged 
violations of treaty). 



REFORMING ASYLUM ADJUDICATION 237 



resulted, frequently marked by interventionist judicial rulings and 
unwise government reaction.^ 

In late 1988, these controversies returned to the front pages. 
Whereas this country received about 3000 asylum applicants per 
year in the 1970s, by December 1988, the Immigration and Naturah- 
zation Service (INS) was receiving 2000 applicants per week in South 
Texas alone, nearly all of them from Central America. Neither the 
basic paperwork, nor even minimally decent food and shelter could 
keep pace. Some commentators concentrated on the humanitarian 
plight of the individuals. Others worried about the burdens placed 
on state governments and local communities, especially since they 

s Class-action suits over Haitian asylum seekers presented the first major 
challenges to the government's procedures for asylum adjudication and ultimately 
resulted in years of litigation, blocking return of nearly all affected persons. The first 
round, begun in 1975, focused on procedures for considering asylum claims lodged 
by excludable aliens. (Excludable aliens are those apprehended at the border before 
making an entry into the United States. When the government seeks to remove 
aliens who have entered, it must do so through deportation rather than exclusion 
proceedings.) The litigation history, including citations to the several reported 
decisions that resulted, is summarized in Sannon v. United States, 631 F.2d 1247, 
1249-50 (5th Cir. 1980). The second round derived from a 1977-78 campaign by the 
Immigration and Naturalization Service (INS) to hasten removals of Haitian asylum 
applicants. Ultimately this "Haitian Program" was declared unlawful, and the court 
ordered INS to reprocess all class members' asylum claims. See Haitian Refugee 
Center v. Smith, 676 F.2d 1023 (5th Cir. Unit B 1982). Moreover, the plaintiffs 
received a major attorneys' fee award under the Equal Access to Justice Act. See 
Haitian Refugee Center v. Meese, 791 F.2d 1489 (lldi Cir.), vacaUd in part on other 
grounds. 804 F.2d 1573 (11th Cir. 1986). 

In 1981, the Administration embarked on a new detention policy meant largely 
to deter asylum applications. See T. Alxinikoff 8c D. Martin, Immigration: Process 
AND PoucY 722-24 (1985). Litigation challenging the detention policy went all the 
way to the Supreme Court in Jean v. Nelson, 472 U.S. 846 (1985). Although the 
government won some impK>rtant victories with regard to legal doctrine, the plaintiffs 
prevailed sufficiently on their statutory and regulatory claims to delay asylum 
processing for lengthy periods, see, e.g., Louis v. Meissner, 532 F. Supp. 881, 883-84, 
889 (S.D. Fla. 1982) (refusing to dismiss refugees' claim that they were denied access 
to counsel contrary to INS regulations); Louis v. Nelson I, 544 F. Supp. 973, 993-96 
(S.D. Fla. 1982) (holding that detention policy was illegal because it was a substantive 
rule that was not promulgated as required by the Administrative Procedure Act); 
Louis V. Nelson II. 544 F. Supp. 1004 (S.D. Fla. 1982) (holding that detained Haitians 
were entitled to release on parole pending determination of their claims for 
admission), and ultimately to gain a sizable attorneys' fee award. See, e.g., Jean v. 
Nelson. 863 F.2d 759. 769-80 (11th Cir. 1988). cert, granted sub nom. INS v. Jean. 58 
U.S.L.W. 3467 (U.S.Jan 23, 1990) (No. 89-601). Similar challenges to INS handling 
of Salvadoran and Guatemalan claims have likewise won initial successes. See, e.g., 
Orantes-Hemandez v. Meese. 685 F. Supp. 1488. 1511 (CD. Cal. 1988) (granting 
permanent injunction); Orantes-Hemandez v. Smith. 541 F. Supp. 351. 380 (CD. 
Cal. 1982) (granting preliminary injunction); Nunez v. Boldin, 537 F. Supp. 578. 587 
(S.D. Tex.) (also granting prehminary injunction), appeal dismissed, 692 F.2d 755 (5th 
Cir. 1982). 



238 DAVID A. MARTIN 

believed that most of the apphcants were abusing the system and that 
their success would only encourage more to come.** In response, 
INS quickly sought ways to deter new arrivals. Despite lawsuits and 
temporary restraining orders, by early spring 1989, INS imple- 
mented a policy providing for initial asylum decisions within one day 
of application, coupled with detention of all unsuccessful applicants 
in South Texas.* 

This action sharply reduced the concentrated influx, eased local 
alarms, and thereby succeeded in draining off much media interest. 
One might almost have thought that the annual application level 
dropped back to the politically tolerable mid-decade rate of around 
20,000 per year. It did not. TTie inflow remained exceptionally high, 
but simply spread more widely throughout the country. In fiscal year 
1989, the district offices of INS received over 100,000 applications — 
a record — and many thousands more were filed before immigration 
judges.® This pace continues. It would not be surprising to find the 
issue bursting into prominence again in the early 1990s, as the fiill 
extent of the new asylum caseload becomes apparent. 

In nearly all Western countries, the asylum adjudication systems 
now employed were cobbled together in an era that permitted lei- 
surely consideration of modest caseloads. In general, they have 
adapted poorly to an era when claims are numerous and subject to 
sudden escalation. Moreover, because most Western adjudication 
systems were built on the rough assumption (a product of the Cold 
War) that few claimants would be rejected, they avoided difficult 
questions about effective information-gathering and evaluation. 
Today's dilemmas require instead a sustained and sophisticated 
capacity to screen out unqualified applicants; hence, the difficult, 
previously latent questions have become inescapable. If adjudication 



* See, e.g., Suro, AUens Stty New Curbs Won't Hall Them at Border, N.Y. Times, Feb. 
27. 1989. at A14, col. 1; Suro. U.S. Set to Detatn Refugees in Tents Beginning Today, N.Y. 
Times. Feb. 21. 1989. at Al, col. 4; Schmalz. Miami, Saymg It's Overburdened, TeUs 
Nicaraguans to Stay Away, N.Y. Times. Jan. 14. 1989. at Al. col. 3; Applebome, South 
Texans Fear an Influx of Aliens, N.Y. Times. Dec. 16. 1988. at A22. col. 1. 

5 See, e.g.. South Texas Asylum Applicants Face Detention Under New INS Policy, 66 
Interpreter Releases 217 (1989); Frelick. South Texas Detention Plan Goes Forward, 
Refugee Reports. Feb. 28. 1989. at 1 (discussing the new INS plan to restrict 
allegedly frivolous asylum claims and the resulting criticism from private agencies); 
Applebome. yu^ff Halts Rule Stranding Aliens m Rio Grande Valley, N.Y. Times, Jan. 10, 
1989. at A 14. col. 1; Applebome. Smt Attacks Policy That Keeps Aliens m Texas, N.Y. 
Times, Jan. 7, 1989. at 1. col. 3. 

^ See infra Table I following note 150. Immigration judges, originally known as 
special inquiry ofhcers, are the federal officials who preside over deportation and 
exclusion proceedings. See T. Aleinikoff Sc D. Martin, si^a note 3, at 87-91. 



REFORMING ASYLUM ADJUDICATION 239 

systems are to say no to large numbers of applicants — as virtually all 
Western countries are now doing — they must either cultivate callous- 
ness to the risk of returning true refugees, or else demand assurance 
that their outcomes are precise and reliable. 

The search for effective reforms continues, but in a highly 
polarized environment. In this Article, I attempt to advance that 
search, with special attention to the political asylum adjudication sys- 
tem of the United States. The analysis and recommendations offered 
here are based on several years of research and observation in the 
United States, Canada, and several European countries — including 
attendance at numerous hearings and extensive interviewing of par- 
ticipants in asylum processing systems, principally government offi- 
cials, private attorneys, and other workers with refugee assistance 
and advocacy organizations.^ 

I open with a general look at the basic legal standards and then 
at the highly charged policy context in which all asylum decisions are 
currendy made. To lay a foundation for the recommendations at the 
end, there follows an overview of the United States* current, tangled 
adjudication procedures and a brief history of past procedures, 
explaining how the present system evolved. I conclude by sug- 
gesting numerous changes in structure and process, some of them 
quite ambitious, while rejecting others that have been prominendy 
advocated. I offer these proposals with the conviction that a well- 
functioning adjudication process provides the indispensable ingredi- 
ent for alleviating the many ills now attributed to the system, even 
though the various participants in the current debate tender widely 
differing diagnoses of what are the most serious ailments. 

I. The Substantive Legal Framework 

A. International Provisions 

Classically, the right of asylum under international law belonged 
to states and not to individuals.^ Sovereigns were considered to 
have the right or prerogative to grant protection against return to 

■^ The longer version of this study, forthcoming from the Yale University Press, 
contains considerably more detail on the Canadian and European systems. See supra 
note f. I will refer directly to those systems only occasionally here, but the 
suggestions I make draw in large measure on what I learned in studying them. 

^ See G. GooDwiN-GiLL, The Refugee in International Law 101-02 (1983); 2 
A. Grahl-Maosen, The Status of Refugees in International Law 6 (1972); 
Melander, Refugees and International Cooperation, 15 Int'l Migration Rev. 35, 35 
(1981). 



240 DAVID A. MARTIN 

those they chose to shelter. This framework shamed itself in the 
world's woefully inadequate response in' the 1930s and 1940s to 
those who were fleeing Nazi persecution. From the ashes of World 
War II arose an international structure that signalled a determina- 
tion, measured but genuine, to do more for refugees. 

Most enduring of the post-war measures were two instruments 
adopted under United Nations auspices. The first, accepted in late 
1950, created a new post of UN High Commissioner for Refugees 
(UNHCR). Initially expected to be temporary, the Office of the 
UNHCR, staffed by approximately 2300 employees, has by now 
become a permanent fixture on the international scene.^ Under its 
original statute and subsequent General Assembly resolutions, the 
office bears responsibility for providing protection and material 
assistance to refugees throughout the world. '^ In connection with its 
protection function, UNHCR monitors asylum adjudication systems 
worldwide, and occasionally plays a direct role in individual 
determinations. 

The second legal instrument, the 1951 Convention Relating to 
the Status of Refugees,** remains of surpassing importance, for it 
established a definition that has become the centerpiece of most 
Western asylum adjudication systems, including that of the United 
States. Under that definition, as improved by the Convention's 1967 
Protocol,** a refugee is a person outside her home country, unwill- 
ing or unable to return or otherwise claim that country's protection 
because of a "well-founded fear of being persecuted for reasons of 
race, religion, nationality, membership of a particular social group or 
political opinion."*' 

9 See Overview of UNHCR Activities: Report for 1987-88, at 87. U.N. Doc. A/AC.96/ 
709 (1988) (Table DC). 

10 See Statuu of the Office of the UNHCR. Gj\. Res. 428. 5 U.N. GAOR Supp. (No. 
20) at 46, U.N. Doc. A/1775 (1950). Later changes in the scope of UNHCR 
responsibility are traced in Aga Khan, Leged Problems Relating to Refugees and Displaced 
Penons. 1976(1) Recueil des Cours 287. 301-09. 

11 July 28. 1951. 189 U.N.T.S. 1S7 [hereinafter 1951 Convention] (adopted in 
Geneva). 

•2 Protocol Relating to the Sutus of Refugees. Jan. 31. 1967, 19 U.S.T. 6223, 
T.I.A.S. No. 6577, 606 U.N.T.S. 267 [hereinafter 1967 Protocol]. The 1951 
definition covered only those persons who feared persecution "[a]s a result of events 
occurring before 1 January 1951." 1951 Convention, ju^a note 11, art. 1(A)(2), 189 
U.N.T.S. at 152. The main function of the 1967 Protocol was to remove this dateline. 
The United States became a party to the Protocol in 1968, thereby becoming 
derivatively bound by all the significant operative provisions of the 1951 Convention. 
See 1967 Protocol, supra, art. 1(1). 19 U.S.T. at 6225. T.I.A.S. No. 6577. at 2. 606 
U.N.T.S. at 268. 

>» 1951 Convention, supra note 11. art. 1(A)(2). 189 U.N.T.S. at 152. The 



REFORMING ASYLUM ADJUDICATION 241 

The 1951 Convention, a cautious and more limited treaty than is 
often appreciated,'^ provides relatively few actual guarantees to ref- 
ugees illegally present in the country of haven (as most asylum seek- 
ers now are).'^ In particular, it does not guarantee asylum, in the 
sense of a durable lawful residence status, even for those duly 
adjudged to be refugees under its provisions. Thus even today there 
is no individual right of asylum under international law.'^ What the 
1951 Convention does require, however, even for refugees illegally 
present, is nonrefoulement — a technical term for protection, deriving 
from Article 33 of the Convention, against return to a country 
"where [the refugee's] life or freedom would be threatened on 
account of her race, religion, nationality, membership of a particular 
social group or political opinion."'^ Article 33 affords a limited and 
country-specific protection, and the receiving nation technically 

definition also contains "cessation" and "exclusion" clauses that remove certain 
persons, such as those guilty of war crimes, or those who have taken on a new 
nationality, from the Convention's coverage. See id. art. 1(C)-(F), 189 U.N.T.S. at 
154-56. 

»-» See Haitian Refiigee Center v. Gracey, 809 F.2d 794, 840-41 (D.C. Cir. 1987) 
(Edwards, J., concurring and dissenting) (concluding that the Refugee Act's 
applicability to aliens in both exclusion and deportation proceedings is more 
generous than required by the 1951 Convention and 1967 Protocol). 

'^ The 1951 Convention does provide a host of impressively detailed 
guarantees for refugees lawfully present, but a decision that the person is a 
Convention refugee does not ipso f^cto result in lawful presence. See Report of the Ad 
Hoc Comm. on SuUeUssness and Related Problem 47, U.N. Doc. E/1618/Corr. 1; £/ 
AC.S2/5/Corr.l (1950). See generally Weis, The International Protection of Refugees, 48 
Am. J. Int'l L. 193, 193-207 (1954) (outlining the protection provisions of the 1951 
Convention). The major purpose of the Convention, as the name suggests, was to 
clarify questions of status for the World War II refugees already in place. As 
Professor Goodwin-Gill has explained: 

The 1951 Convention was originally intended to establish, confirm or 
clarify the leg^ status of a known pK>pulation of the displaced. This met 
the needs of the time, and most provisions focus on assimilation, or are 
premised on lawful residence or tolerated presence. There is nothing on 
asylimi, on admission, or on resetdement. 
Goodwin-Gill, The Future of International Refugee Law, Refugees. Oct. 1988, at 27, 28. 

1^ An abortive effort was made in the 1970s to draft a convention that would go 
further toward international legal guarantees of political asylum for refugees. This 
effon was abandoned when a 1977 conference of government representatives 
appeared likely to weaken even those minimal guarantees derived from the 1951 
Convention and the 1967 Protocol. See Weis. The Draft United Nations Convention on 
Territorial Asylum. 50 BRrr. Y.B. Int'l L. 151. 159, 169-71 (1979). 

>' 1951 Convention, sufrra note 11. art. 33(1). 189 U.N.T.S. at 176. Paragraph 
(2) of this article authorizes narrow exceptions to the nonreturn obligation, 
essentially for spies and dangerous criminals. 



242 DAVID A. MARTIN 

remains free to send a refugee on to other countries, rather than 
granting asylum on its soil.'® 

Nevertheless, since 1951 most Western countries, to their 
credit, have set up asylum claims systems that essentially combine 
the determination of refugee status under the 1951 Convention defi- 
nition with the discretionary act of providing durable status, or asy- 
lum. An affirmative refugee status determination thus routinely 
leads not only to the limited protection against return contemplated 
by Article 33, but also to the full range of protections embraced 
within the notion of asylum.'^ In this sense, we have come close to a 
system that guarantees an individual right of asylum to those who 
somehow establish physical presence on the soil of such Western 
countries and also prove that they satisfy the 1951 Convention 
definition. 



>B See Melander. supra note 8, at 36. The country is also free, of course, to grant 
asylum to others it deems worthy, even if they do not satisfy the 1951 Convention 
definition. Western European countries have done this more extensively than the 
United States, through so-called "B-status" refugee provisions or the acceptance of 
"de facto refugees." See generally D. Gauacher, S. Martin & P. Facen, Temporary 
Safe Haven: The Need for North American-European Responses (Refugee Policy 
Group 1987). The United States occasionally provides such protection, without a 
determination of refugee status, through the use of "extended volunury departure" 
(EVD). See generally T. AtfiNiKOFF Sc D. Martin, si^na note 3, at 726-43 (explaining 
EVD protection and the debates surrounding it); Note, Temporary Safe Haven for De 
Facto Refugees from War, Violence and qisasters, 28 Va. J. Int'l L. 509, 512 (1988) 
(arguing for the adoption of alternative methods of offering a temporary safe haven 
rather than the currendy inadequate EVD system). Under current law, EVD is 
provided as a matter of grace by the political branches, according to ad hoc criteria 
that are essentially beyond the reach of judicial review. See Hotel and Restaurant 
Employees Union Local 25. v. Smith. 594 F. Supp. 502 (D.D.C. 1984). c^'d by an 
equally divided court, 846 F.2d 1499 (D.C. Cir. 1988); see also Foreign Relations 
Authorization Act, Fiscal Years 1988 and 1989, Pub. L. No. 100-204, { 902, 101 Sut. 
1331 (1987) (esublishing limited procedure for eventual permanent resident status 
for certain beneficiaries of EVD for the years 1982-87). 

Although I believe that this country should extend such protection on carefully 
chosen occasions to potential victims of civil strife or human rights violations, I do 
not address here either the standards or procedures for such decisions. This study 
focuses instead on what are supposed to be nonpolitical procedures for 
implementing binding, neutral criteria adopted to shield those most seriously 
jeopardized by granting them fiill asylum or, at a minimum, nonrefoulement. 

•9 See Hofinann, Asylum and Refugee Law, in The Legal PosmoN of Auens in 
National and International Law 2045. 2058-59 (J. Frowein Sc T. Stein eds. 1987). 
In fact. Western nations (with a few exceptions, like Austria, traditionally viewed as 
transit countries) rarely find third countries willing to take refugees off their hands. 
Given that the refugees are present and. under Article 33. cannot be sent to the only 
country normally obligated to take them (the country of nationality), it is clearly 
better if they quickly atuin a secure new status that allows them to rebuild normal 
lives. See Martin, supra note 1, at 18 n.26. 



REFORMING ASYLUM ADJUDICATION 243 

That these admirable features of the system go beyond the strict 
requirements of international law, however, should remind us of 
their fi-agility. They cannot be taken as inevitable constants. 
Instead, it must be an ever-present concern of wise policy to shape 
asylum measures, including adjudication systems, so as to maximize 
continued domestic support. The systems' inability to cope effec- 
tively with growing numbers of asylum seekers over the last decade 
now threatens that foundation.^® 

B. American Legal Proxnsions 

The American leg^ framework follows these general oudines, 
but the details require some additional attention. Although the 
United States played a significant role in the conferences that led to 
adoption of the final text of the 1951 Convention, this country never 
became a party to that treaty. During that era, bitter batdes over the 
Genocide Convention and the Bricker Amendment had resulted in 
ill-considered executive promises against sending any human rights 
treaties to the Senate for ratification.^' Nevertheless, the United 
States regarded itself as a leading player in finding solutions to refu- 
gee problems. From the end of World War II, under a variety of 

^ A recent book by a former UN Deputy Hig^ Commissioner for Refugees 
underscores these risks. Richard Smyser writes: 

The structure of refugee law and care, which has been generously 

assembled since the dawn of our culture and particularly in the twentieth 

century, cannot remain in place if it is abandoned by political and popular 

opinion. If the people of the world decide that they no longer wish to 

receive and help refugees, all the international conventions and 

organizations will be rendered useless and will prove unequal to the task 

of saving even a single life. That is a danger that must be averted. 

W. Smyser, Refugees: Extended Exile 2 (1987); see also Martin, supra note 1, at II- 

15 (noting that "[p]ohcy initiatives that will impiose costs and difficulties on those 

publics in a highly visible way must ultimately be grounded in a widespread public 

acceptance of the need for that policy"). 

21 See generally Kaufinan &: Whiteman, Opposition to Human Rights Treaties in the 
United States SenaU: The Legacy of the Bricker Amendment, 10 Human Rights Q. 309, SIO- 
12 (1988) (discussing the Bricker Amendment and the executive branch's response). 
Public explanations of the U.S. position regarding the refugee treaty usually focused, 
with some justification, on how the 1951 Convention better fit European leg^ 
systems, where the status of an alien depended on reciprocity — unlike the United 
States, where refugees received a status equivalent to other permanent resident 
aliens. See Hoffman, UN Refugee Code Near Completion: WcUered-Doum BUI of Rights Applies 
Mainly to Europe— Some Basic Gains Made, N.Y. Times, July 21, 1951, at 4, col. 2. At 
other times American spokesmen stated that this country would not adhere because 
the treaty touched largely on matters within the jurisdiction of the states rather than 
the federal government. See Eleven Nations Sign Refugee Rights Act, N.Y. Times, July 29, 
1951. at 15, col. 3. 



244 DAVID A. MARTIN 

Statutory and administrative schemes, it had generously resettled 
hundreds of thousands of the displaced persons uprooted by that 
conflict.^^ That experience imprinted on American policy debates a 
distinctive perspective that predominated until quite recently: 
Responding to refugees meant resettling displaced persons from ref- 
ugee camps overseas, rather than dealing with populations already 
on national territory.^^ 

Even during this period, however, some provision was made for 
the handful of individuals who somehow made it to American terri- 
tory on their own and then asked not to be returned to face persecu- 
tion. Congress enacted the first express statutory provision in 
1950,^'* directing the Attorney General not to deport aliens to coun- 
tries where they "would be subjected to physical persecution."** 

In a more explicitly discretionary form, this provision was incor- 
porated as section 243(h) of the newly codified Immigration and 
Nationality Act (IN A) in 1952. It granted the Attorney General the 
discretion to "withhold deportation" of persons who would be sub- 
jected to physical persecution upon return.*® In 1958, the Supreme 
Court ruled that this statutory provision applied only to deportation 
and was not available to aliens in exclusion proceedings,*^ but the 
INS made equivalent protections available to excludable aliens 

22 A comprehensive account appears in Congressional Research Library for 

THE COMMrrTEE ON THE JUDICIARY OF THE UNfFED STATES, 96tH CoNC, 2d SeSS., 

Review of U.S. Refugee Resettlement Programs and Poucies (Comm. Print 
1980). For a more critical review, see G. Loescher &: J. Scanlan, supra note 2, at 1- 
67. 

25 See Meissner. Reflections on the Refugee Act of 1980, in The New Asylum 
Seekers: Refugee Law in the 1980s 57. 60 (D. Martin ed. 1988). Sharply different 
policy constraints o(>erate in the two settings, despite similarities in the governing 
threshold standards. See infra note 38. 

2'* Related provisions had appeared, however, since 1875. They provided an 
exception to exclusion based on pre-entry conviction of a criminal offense, if the 
crime constituted a pohtical offense. See T. Aleinikoff & D. Martin, supra note 3, at 
638&:n.lS. 

25 Subversive Activities Control Act of 1950, ch. 1024. % 23, 64 Stat. 987. 1010 
(repealed 1952). 

26 Immigration and Nationality Act. ch. 477. § 243(h). 66 Sut. 163. 214 (1952) 
(codified at 8 U.S.C. § 1253(h) (1988)). In 1965. Congress deleted the reference to 
"physical persecution," replacing it with "persecution on account of race, religion, or 
pohtical opinion." Act of Oct. 3. 1965. Pub. L. No. 89-236. § 1 1(0. 79 Stat. 911.918. 
The Refugee Act of 1980. Pub. L. No. 96-212, § 203(e). 94 Stat. 102, 107. gave the 
section its present shape, more closely tracking the language of Article 33 of the 1951 
Convention. 

27 5*^ Leng May Ma v. Barber, 357 U.S. 185 (1958). 



REFORMING ASYLUM ADJUDICATION 245 

through the use of the parole power, likewise an expressly discre- 
tionary remedy.^® 

By 1968, the earlier resistance to human rights treaties had soft- 
ened sufficiendy for the Johnson Administration to send the 1967 
UN Protocol to the Senate, where it secured speedy ratification. 
Because the 1967 Protocol incorporates by reference all of the 
important operative provisions of the 1951 Convention, with one 
important modification in the definition of "refugee,"^^ ratification 
was tantamount to acceding to the earlier instrument. But this some- 
what circuitous route toward accepting the 1951 obligations appar- 
endy helped avoid reopening any of the previous decade's treaty- 
power battles. 

The Administration had promoted the 1967 Protocol primarily 
as a way of signalling U.S. leadership on worldwide refugee issues 
and encouraging other nations, regarded as less supportive of refu- 
gees, to improve protections. For that reason, political asylum issues 
drew litde attention during the Senate's brief deliberations on the 
treaty.*** An unexamined assumption that U.S. practices conformed 
fully to the 1951 Convention's requirements permeated the proceed- 
ings, and executive spokespersons assured the Senate that the 1967 
Protocol could be implemented without changes in the statutes. 
Although this was true, the record suggests that the Senate probably 
did not fully appreciate the significance of the treaty with respect to 
the withholding of deportation. After accession to the 1967 Proto- 
col, the United States came under a firm legal obligation to imple- 
ment § 243(h), a discretionary provision, so as not to conflict with 
the mandatory requirements of Article 33 of the 1951 Convention, 
the nonrefoulement provision. In any event, the treaty deliberations 
clearly did not provoke consideration of any difficult issues concern- 
ing the substantive legal provisions or the adjudication procedures 
that would be used to implement them. No changes were made at 
the time in the substantive statutory requirements affecting political 
asylum. 

There matters stood until Congress considered the bills that 



28 See, e.g., 8 C.F.R. §§ 236.3(e), 253.1(f) (1978) (referring to limited circum- 
stances in which otherwise excludable refugee aliens may benefit from parole). The 
general authorization for parole appears in INA § 212(d)(5). 8 U.S.C. § 1182(d)(5) 
(1988). See generally T. Aleinikoff ic D. Martin, supra note 3, 232-36. 

29 See supra note 12. 

»o See S. Exec. Doc. K, 90th Cong., 2d Sess. (1968) (transmitul materials from 
executive branch); S. Exec. Rep. No. 14. 90th Cong., 2d Sess. (1968) (Committee 
report on the 1967 Protocol, including a transcript of the hearings). 



246 DAVID A. MARTIN 

were to become the Refugee Act of 1980. The political branches 
took up refugee issues at that time primarily because of the difficul- 
ties encountered in coping with the massive refugee outflows from 
Indochina— classic overseas refugee issues." Even though asylum 
applications were increasing throughout the period of legislative 
deliberation and a significant political and judicial controversy was 
brewing in Florida regarding Haitian asylum seekers, asylum was 
ag^in largely a legislative afterthought. Nevertheless, Congress 
made some important improvements in the asylum realm, urged on 
by UNHCR and by activists who were becoming more vocal about 
domestic asylum issues. First — a matter of particular UNHCR con- 
cern — Congress changed INA § 243(h) to a mandatory form, leaving 
no doubt about the obligatory character of the nonrefoulement provi- 
sions in domestic law, and specifying that the provision applies to 
both exclusion and deportation.'^ Second, Congress finally added 
an express "asylum" provision to the INA, in the form of a new 
§ 208.'' It replaced earlier haphazard administrative practice with a 
new, express, and clarified immigration status for those recognized 
as refugees after applications filed in this country. 

Section 208 states that the Attorney General has discretion to 
grant asylum to aliens who meet the definition of refugee provided in 
the new INA § 101(a)(42)(A).** That section, in turn, tracks the 
1951 Convention definition; the central qualification is a "well- 
founded fear of persecution" on account of the same five factors 
listed in the Convention." The new immigration status, called 
"asylee" in the regulations, clarifies the alien's entidements to cer- 
tain benefits in this country. It also enables him, after a minimum of 
one year as an asylee, to obtain full lawful permanent resident status 

" See Anker ic Posner, The Forty Year Crisis: A Legislative History of the Refugee Act of 
1980. 19 San Diego L. Rev. 9. 30 (1981); Martin, The Refugee Act of 1980: Its Past and 
Futwe, in Michigan Yearbook of International Legal Studies: Transnational 
Legal Problems of Refugees 91, 92-96 (1982). 

*2 It now provides: "The Attorney General shall not deport or return any alien 
... to a country if the Attorney General determines that such alien's life or freedom 
would be threatened in such country on account of race, religion, nationality, 
membership in a particular social group, or pohtical opinion." 8 U.S.C. § 1253(h)(1) 
(1988). 

" 5« 8 U.S.C. § 1158 (1988). 

^ See id.: see also 8 U.S.C. §1101(a)(42) (1988) (for the new INA 
§ 101(a)(42)(A)). 

S5 See id. 



REFORMING ASYLUM ADJUDICATION 247 

through a statutorily authorized adjustment procedure.'^ No such 
reg^ar adjustment procedure for asylees existed theretofore.'^ 

Again Congress paid Httle attention to details of adjudication 
procedures or substantive standards,'® but one theme is clear from 

*^ SeeS C.F.R. pts. 208. 209 (1989). The statute imposes an annual ceiling of 
5000 on adjustments of asylees to permanent resident status. See 8 U.S.C. i 1159(b) 
(1988). This is a ceiling on adjustments only; it does not limit the number of people 
who may receive aisylum in a given year. See generally Martin, sttpra note 3 1 . Because 
asylum grants have nm considerably above 5000 for the last several years, a backlog 
has developed and asylees must now wait much longer than one year before 
adjusting and thus receiving a "green card." See December Visa Numbers Move Only 
Slightly. 66 Interpreter Releases 1269-70 (1989). 

'7 This statement must be qualified in one minor respect. The principal 
provision expressly meant for refugees from 1965 to 1980 was INA § 203(a)(7), 8 
U.S.C. i 1153(a)(7) (1976). It allowed the use of six percent of the "preference" 
admission spaces each year for people who fled persecution in communist countries 
or the general area of the Middle East. The overwhelming majority of these refugee 
spaces (a total of 17,4(X) annually in the late 1970s) were used for overseas refugee 
programs, principally for the admission of Eastern Europeans selected in Western 
Europe, suid in the Late 1970s for Indochinese processed in Southeast Asia. See T. 
Aleinikoff & D. Martin, suf/ra note 3, at 620-23. 

A statutory proviso to § 203(a)(7), however, permitted use of a portion of these 
admission spaces to grant permanent resident status to aliens who met the basic 
requirements but had been physically present in the United States for at least two 
years before applying for such adjustment of status. See 8 C.F.R. § 245.4 (1971). For 
example, a Polish national who overstayed a nonimmigrant visa could receive this 
status through a petitioning process carried out entirely in this country, if she proved 
the requisite two years' presence and flight fi-om persecution. In a sense, she thus 
could be considered an asylee who graduated to full inmiigrant status under a 
permanent legislative provision. But the term "asylee" was never formally applied to 
the status of those who benefited from this proviso, and successful applicants moved 
directly to permanent resident status, usually from some sort of irregular status. 
Moreover, the qualifying standards departed somewhat from the provisions of the 
1951 Convention (most graphically in the geographic limitations). 

'8 Congress also applied the UN refugee definition to overseas refugee 
programs under INA § 207, 8 U.S.C. § 1 157 (1988), which usually operate by way of 
INS interviews and screening in refugee facilities in third countries, such as Thailand 
or Austria. Because of this similarity in qualifying standards, one might assume that 
the system proposed here for asylum adjudications should therefore be applied to 
"refugee" adjudications in the overseas prog^ram. I would argue against such a 
conclusion. Tlie widely different functional constraints operative in the overseas 
program counsel ag^nst identical determination systems. 

First, simply because of physical location, the United States is able to apply 
numerous other screening criteria, as well as numerical ceilings, before deciding 
which refugees will be offered resettlement in the United States as part of the 
overseas programs. Historically, screening and selection for these programs have 
been based most importantly on these other criteria, such as family or other ties in 
the United States, rather than on satisfaction of the refugee definition. Therefore, 
pouring extensive resources into the adjudication of the latter issue often is not 
advisable. Second, in most such overseas circumstances, a decision to exclude 
certain applicants from the U.S. refugee program, on whatever grounds, does not 
necessarily mean return to their home country. Typically, such persons remain the 



248 DAVID A. MARTIN 

the legislative history. Congress intended the refugee standards to 
be applied neutrally and without ideological bias, in contrast to cer- 
tain repealed refugee provisions that had made special provision for 
persons fleeing Communist countries.'^ Although occasional argu- 
ments have appeared, particularly during the Reagan Administra- 
tion, for a more overtly political selection system,*** neutral 
application represents by far the fairer and wiser policy. In the long 
run, favoring some groups in the asylum process only increases the 
political costs of returning other individuals, even when their claims 
are accurately rejected under an appropriately demanding application 
of the governing standards. Full consideration of this comphcated 



responsibility of the first-asylum country, which may be able to find other 
resettlement opportunities for them. 

The procedures suggested here are crafted for the sharper choices faced in U.S. 
asylum processing, wherein the government has essentially only two options once a 
person is adjudged a refugee, owing to the person's presence on U.S. territory and 
the usual unwillingness of third countries to contribute resettlement spaces: grant 
asylum (or at a minimum nonrefotUement) and allow indefinite stay here or deny 
protection and return the applicant to the home country. In that setting, greater 
assurance of accuracy and professionalism in applying the refugee definition is 
essential. I have elaborated on these distinctions in Martin, supra note 31, at 1 1 1-14. 

S9 See Anker & Posner, si^a note 31. at 12, 14-18, 41-43, 60, 64. Most of the 
congressional statements criticizing the earlier "discriminatory" provisions 
specifically addressed the overseas refugee program, because asylimi provisions 
received little attention. There is no reason to doubt, however, that Congress 
expected the same neutral application in asylimi processing, where the case for stria 
but evenhanded application of the refugee definition is probably Ear stronger. See 
Martin, supra note 31, at 1 13-14. 

^ In 1986, the Justice Department under Attorney General Meese was 
reportedly drafting new asylum regulations that would have estabUshed a 
presumption in favor of those fleeing "totalitarian" countries — apparently including 
all Communist countries. See Pear, Plan to Give More Poles Asylum is Under Study by 
Administratum, N.Y. Times, Mar. 30, 1986, § 1, at 1, col. 4. No such regulations ever 
appeared. In 1987, Mr. Meese did announce a set of steps relating to Nicaraguans. 
Although she was under pressure from some conservative circles hostile to the 
Sandinista government to grant blanket permission to stay ("extended voluntary 
departure") to all Nicaraguans, her announcement nominally only restated 
established standards for ordinary asylum determinations. In practice, however, that 
statement encouraged more Nicaraguans to come forward and apply, and it has led 
to a far higher grant rate for Nicaraguans applying for asylum in INS district offices. 
Special review by the central office in Washington was also required before any 
Nicaraguan was deported. As a result, such deporutions became extremely rare, 
even when asylum was denied. See Refugee Reports, July 10, 1987, at 7-8; Refugee 
Reports, Aug. 14. 1987, at 8-10. 

For broader consideration of the merits of a more expressly political refugee 
program (arguments that carry more weight with respect to an overseas refugee 
progn^m rather than an asylum system), see M. Walzer, Spheres of Justice: A 
Defense of Plurausm and Equauty 49-50 (1983); Suhrke, Global Refugee Movements 
and Strategies of Response, in U.S. Immigration and Refugee Poucy: Global and 
Domestic Issues 157. 157-62 (M. Kriu ed. 1983). 



REFORMING ASYLUM ADJUDICATION 249 

debate is beyond the scope of this study. Nevertheless, the study is 
premised on the assumption that Congress's 1980 approach is to be 
continued. It concentrates on finding effective ways to implement a 
neutral system meant to protect those most seriously at risk of perse- 
cution, whatever the poUtical orientation of the home-country 
government. 

In the early years of asylum adjudication under the Refugee Act, 
immigration authorities and refugee advocates alike assumed that 
the threshold standard for applying sections 208 and 243(h) was 
identical.** But the Act revived litigation over the precise under- 
standing of that standard. The Board of Immigration Appeals (BIA), 
the body that hears appeals from immigration courts,**^ had tradi- 
tionally required the applicant to show *'a clear probability of perse- 
cution" in order to gain withholding of deportation under the pre- 
1980 version of § 243(h), whereas refugee advocated had consist- 
endy urged the adoption of some more generous standard.** When 
the Board declined to change its "clear probability" formula after 
enactment of the Refugee Act, numerous applicants challenged the 
rulings in the appellate courts.** The circuits divided on the ques- 
tion, and the issue reached the Supreme Court in 1984 in INS v. 
Stevic.*^ That Supreme Court ruling unexpectedly split the qualify- 
ing standards. It ruled that the Board's traditional "dear 
probability" test still governs in order to claim the mandatory pro- 
tection of § 243(h), while hinting, without expressly ruling, that 
some easier standard might apply under § 208.*® Stevic, however, 
did provide a softening gloss on the Board's "clear probability" test, 
reading it to require a showing only that persecution is "more likely 
dian not."*' 



*• See. e.g., Helton, INS v. Cardoza-Fonseca: The Decision and its Implications, 16 
N.Y.U. Rev. L. & Soc. Change 35, 39 n.31 (1987-88); Scanlan, supra note 2, at 625. 

*^ See ir^a notes 171-174 and accompanying text (discussing the Board of 
Immigration Appeals). 

-»» See, e.g., Rosa v. INS, 440 F.2d 100, 102 (1st Cir. 1971) ("The burden was . . . 
to prove that there is a dear probability that she will be subjected to persecution if 
deported."); Cheng Kai Fu v. INS. 386 F.2d 750, 753 (2d Cir. 1967) (stating that 
"only where there is a clear probability of persecution ... is this discretion to be 
favorably exercised"), cert, denied 390 U.S. 1003 (1968); In re Dunar, 14 1. &: N. Dec. 
310, 318-19 (1973) (reaffirming "clear probability of persecution" standard after the 
U.S. became a party to the 1967 Protocol). 

♦4 See, e.g., In re McMullen, 17 I. & N. Dec. 542, 544 (1980) (reiterating the 
"clear probability of persecution" standard). 

« 467 U.S. 407 (1984). 

-•6 See id at 430. 

-♦^ Id at 429-30. 



250 DAVID A. MARTIN 

Three years later, in INS v. Cardoza-Fonseca,*^ the Court finally 
ruled squarely on the threshold standard determining eligibility for a 
discretionary grant of asylum under § 208. The majority overruled 
the BIA's continued assertion that the two standards remained iden- 
tical, and forcefully stated that the § 208 test is more generous than 
the standard for § 243(h) .'•^ The Court declined, however, "to set 
forth a detailed description of how the well-founded fear test should 
be applied," leaving that term to acquire "concrete meaning through 
a process of case-by-case adjudication. "*° The § 208 test that has 
come to govern in the wake of Cardoza-Fonseca is most helpfully 
phrased as requiring "a reasonable possibility of persecution," or a 
showing of a "good reason to fear persecution" — but even these 
standards leave much leeway for application to the evidence 
presented in a particular case.*' 

These two Supreme Court cases bring curious results, to say the 
least. To my knowledge, no other country draws this sort of distinc- 
tion between the substantive standards for determining refugee sta- 
tus, on the one hand, and nonrefoulement on the other. *^ Where there 
are distinctions, they typically run in the opposite direction: toward 
shielding more people against return, even if they do not stricdy meet 
the refugee definition and will not be granted the full range of bene- 



48 480 U.S. 421 (1987). 

« S« ii at 425-24. 

*o Id. at 448. 

51 See Inn Mogharrabi. Interim Dec. No. 3028. slip op. (BIA June 12. 1987) 
(implementing the Cardoza-Fonseca ruling and spelling out new guidelines for asylum 
cases). Unfortunately, however. Mogharrabi followed the lead of a Fifth Circuit case, 
Guevara-Flores v. INS. 786 F.2d 1242 (5th Cir. 1986). and restated the sundard as 
follows: "[A]n applicant for asylum has established a well-founded fear if she shows 
that a reasonable person in her circumstances would fear persecution." Mogharrabi, 
slip op. at 9. 

This formulation is misleading and unhelpful. If there is any significant level of 
persecution in a country, a reasonable person would fear becoming its victim, simply 
because of residence in that society, even if the abuses, to date, have never been 
directed at her or persons like her. She may recognize that the chances of actually 
being persecuted are slim, but we surely would not count her out of the realm of 
reasonable persons if she harbors a fear of persecution. In short, a "reasonable 
person" would fear persecution well before the time when we would consider that 
the persecution has become a "reasonable possibility." 

The Board continues to invoke the "reasonable person" standard but appears in 
practice to look for a "reasonable possibility of persecution" — a more objective 
inquiry — before granting asylum. It would promote greater candor in adjudication 
to revert to exclusive use of the "reasonable possibility" formulation, which is, after 
all, the precise wording used by the Supreme Court in dictum in Stexnc and Cardoza- 
Fonseca. 

" See Helton, supra note 41, at 39, 53. 



REFORMING ASYLUM ADJUDICATION 251 

fits that come with formal asylum.*' Moreover, this American bifur- 
cation is subject to substantial objection on policy grounds. It would 
permit American immigration authorities to deny asylum, perhaps 
quite frequendy, in the exercise of discretion. Indeed, the Supreme 
Court in Cardoza-Fonseca went out of its way to emphasize that the 
Attorney General has discretion over these matters, and it plainly 
considered that the holding would increase her "flexibility" in 
responding to refugee crises.** Ostensibly this means that the Jus- 
tice Department could even deport to their home country persons 
already judged to be "refugees" under the Cardoza-Fonseca standard, 
if they fsdl short of the showing required to claim the mandatory 
nonrefouUment protection, as interpreted in Stevic.^^ 

To their genuine credit, the immigration authorities to date 
have avoided any such draconian use of the flexible discretion the 
Supreme Court ratified in Cardoza-Fonseca. The BIA has even moved, 
quite wisely, to limit discretionary denials of asylum and thus provide 
the more complete protection of asylee status for nearly all who meet 
the lower § 208 threshold.*^ As a result of this administrative prac- 
tice, the only important test, in the vast majority of today's asylum 
cases, is the more generous § 208 standard, the "well-founded fear 
of persecution" test. If the alien meets this threshold qualification, 
and is also found worthy of the relief as a matter of discretion (as 
now usually happens), then there is no reason to consider the issues 
raised by § 243(h), because asylum status, by definition, carries with 
it protection against deportation or exclusion.*^ 

*' See, e.g., Goodwin-Gill, Non-RefouUment and the New Asylum Seekers, 26 Va. J. 
Int'l L. 897, 900-02 (1986) (describing these practices). 

M See 480 U.S. at 444-45. 449-50. 

** These points are developed at greater length in T. Aleinikoff & D. Martin, 
supra note 3. at 664-67; id. at 79-80 (Supp. 1987). 

56 See, e.g. In re Pula, Interim Dec. No. 3033. slip op. (BIA Sept. 22. 1987) 
(holding that in the absence of adverse factors, asylum should generally be granted in 
the exercise of discretion). For a comprehensive review of these issues, urging that 
discretionary denials of asylum be used rarely if the individual is found to be a 
refugee (an approach quite similar to that ultimately adopted in Pula), see Anker, 
Discretionary Asylum: A Protection Remedy /or Refugees under the Refugee Act of 1980, 28 Va. 
J. Int'l L. 1 (1987). See also Anker ic Blum, New Trends in Asylum Jurisprudence: The 
Aftermath of the U.S. Supreme Court Decision in INS v. Cardoza-Fonseca, 1 Int'l J. Refugee 
L. 67 (1989); Anker & McGrath, The New Battleground of Asylum Eligibility, in World 
Refugee Survey — 1988 in Review 29 (1989) (arguing that the BIA has begun to take 
a narrower view of what constitutes persecution and has thus made it more difficult 
for refugees from countries in the midst of civil war to gain refuge in the U.S.); 
Helton, The Proper Role of Discretion in Political Asylum Determinations, 22 San Diego L. 
Rev. 999 (1985) (discussing the recent expansion of administrative discretion and 
limits, both domestic and international, on the exercise of that discretion). 

" See In re Mogharrabi, Interim Dec. No. 3028. slip op. at 3 (BIA June 12. 1987). 



252 DAVID A. MARTIN 

Even if the bifurcation of standards someday becomes more 
important in the administrative scheme, it will have no significant 
bearing on the issues of procedural design considered in this study. 
As the Board has recognized, "the core of evidence and testimony 
presented in support of the asylum and withholding applications will 
in almost every case be virtually the same."^® Consequendy, the 
basic process of information gathering and evaluation will not differ, 
whatever the final calibration of the substantive legal standard. In 
principle, adjudicators must in either case first reach a judgment 
about the level of danger faced by the applicant in the home country. 
Only after making that combined factual and predictive assessment 
will they apply the respective legal test in order to determine 
whether to say yes or no to the application for the precise protection 
at issue. In what follows, therefore, reference to "asylum" determi- 
nations should be taken to encompass the adjudication process nec- 
essary to apply the nonrefotdement standards as well. 

II. The Poucy Context 

A. Angles of Vision 
1. The Asylum Tradition 

The commitment to affording asylum to the persecuted is deeply 
rooted in American experience and tradition. TTie Statue of Liberty 
is a treasured icon, perhaps the purest single symbol, in a richly 
diverse nation, of our national self-identity. Furthermore, awareness 
of the grave consequences that may await a refugee wrongfully 
returned to the home country plainly deepens this commitment. No 
successful policy can ignore the instinctively favorable reaction that 
refugees evoke from the American public, politicians, and even (or 
especially) judges. Refugee advocacy groups know that they have a 
ready hold on public imagination, provided only that they can per- 
suade their listeners that the objects of their advocacy are indeed 
refugees. This is as it should be. It is a proud tradition, one that 
should be preserved and strengthened. 

Unfortunately, the very vigor of this tradition carries the seeds 
of difficulty once it is translated into administrative operation. As 
will be seen, accurate asylum determinations require the careful 
application of expertise to a body of information about the individual 
asylum seeker that is, at best, difficult to marshal. But few Americans 

M Id. at 12. 



REFORMING ASYLUM ADJUDICATION 253 

think of this as a job for experts. Few are disposed to defer to the 
judgments of the agency primarily responsible for these decisions,*® 
if the outcomes conflict with their own sense of obligation to 
America's heritage. This attitude accounts, perhaps, for the ambi- 
tious interventionist stance sometimes taken by judges,^ and for the 
impulse toward sudden swings in policy that can come when a new 
set of executive branch officials becomes involved in the process.^' 
It also means that debates on asylum issues often become bitterly 
polarized, for those who oppose deportation of unsuccessful asylum 
applicants often see the matter as a life-or-death issue. 

These attitudes have an important operative significance for asy- 
lum adjudication reform. Any reforms seen as substantially more 
restrictive will face a heavy burden of proof in the public, the media, 
and Congress. Numerous earlier reform efforts have been defeated 
by resistance rallied in those forums by advocacy groups skillfully 
drawing on the Statue-of-Liberty tradition. If stalemate or retrench- 
ment is to be avoided, reforms cannot be done on the cheap. They 
must include ample protections designed to win the support of rele- 
vant domestic audiences (including judges) by giving assurance that 
any new restrictions will not fall unjusdy on deserving asylum 
seekers. 

2. The Need for Control; Asylum as a Loophole 

There is, however, another important angle of vision on the 
promise of asylum, and it prevails with some domestic constituen- 
cies. This oudook derives from the singular trumping power of a 
successful asylum claim. Such a claim overcomes virtually all the 
other qualifying requirements for immigration to the United States. 
It also moves the applicant to the head of the line for early perma- 

*9 This reaction is no doubt fortified by the low esteem that immigration 
agencies usually enjoy within the bureaucratic hierarchy. See M. Morris, 
Immigration — ^The Beleaguered Bureaucracy 87-94 (1985). Sometimes this 
attitude bursts forth in starding fashion. See, e.g., INS v. Cardoza-Fonseca, 480 U.S. 
421, 451-52 (1987) (Blackmun, J., concurring) (leveling harsh criticism at INS— the 
wrong agency (it was the BIA) — in an opinion that takes litde account of the 
complexity of the issues at stake). 

60 See. e.g., Haitian Refugee Center v. Civiletd, 503 F. Supp. 442, 455 (S.D. Fla. 
1980). modxjud sub nom. Haiuan Refugee Center v. Smidi, 676 F.2d 1023 (5di Cir. 
1982). 

6' For example, in her early months as Attorney General, Edwin Meese 
reportedly sought ways to change asylum policy and make asylum more nearly 
automatic for those fleeing Communist countries, as distinguished from those fleeing 
"authoritarian" countries. See N.L. Zucker 8c N.F. Zucker, The Guarded Gate: 
The Reauty of American Refugee Poucy 143 (1987); supra note 40. 



254 DAVID A. MARTIN 

nent residence rights, even if she first estabhshed her presence in the 
territory in knowing violation of the regular provisions of the immi- 
gration laws. Viewed in narrow compass, this too is as it should be. 
Those who have been victimized by persecution should indeed 
receive, early on, a secure new status that will allow them to rebuild a 
new life in a new homeland, without undue insistence on the bureau- 
cratic niceties of ordinary immigration law. 

There are millions of people around the world, however, who 
face no substantizil threat of persecution but who would value such a 
chance at permanent residence in a stable and wealthy nation.^^ In 
an apt simile, Michael Walzer has compared the affluent Western 
democracies to "elite universities, besieged by applicants."®' Lack- 
ing family ties or scarce job skiUs, most of these "applicants" have no 
real prospect of success through any ordinary legal channel. It 
should not be surprising, then, that those who learn about the power 
of a claim to refugee status might choose to try their luck with an 
asylimi application. After all, the only dear requisites for such a fil- 
ing are physical presence on the soil of a Western democracy and 
persistence in asserting the claim. The potential is so promising that 
it has odled into being a new class of entrepreneur "travel agents" 
who arrange for transportation and also instruct their clients on how 
to file for asylum once they encounter officials in the targeted West- 
em country.®* Seen in this light, asylum becomes a major loophole 



^ Lest this sentence be thought unduly ethnocentric or alarmist, one should 
add an imp>ortant qualification. For most such people, their first choice would surely 
be to enjoy stability and prosperity at home, in a political and cultural environment 
with which they are familiar. Indeed, such atuchments to home will doubdess always 
hold most of the population there, even if Western countries suddenly become much 
more hospitable to asylum claims. Alarmist cries suggesting that liberal asylum 
policies may cause whole countries to empty out into a migrant stream to the North 
are thus clearly exaggerated. Cf. Rudge, Don't Blame the Victim, World Link, May 
1988, at 68 (warning against "the apocalyptic scenario suggesting the whole world 
plans to seek asylum" in the West). But, given that stability and pros(>erity are not 
realistic medium-term prospects in many developing nations, it is quite natural that 
some proportion of the population will begin looking elsewhere. And even if the 
proportion is low relative to total home-country (>opulation, the absolute numbers of 
migrants can become sufficiently hig^ to pose a major political problem in the 
receiving state. 

6' See M. Walzer, supra note 40, at 32. 

** See, e.g., E. Ratushny, A New Refugee Status Determination Process for 
Canada 20 (1984) (a report to the Minister of Employment and Immigp-ation, noting 
reports of such agents making arrangemenu for "multiple claimants who have read 
in local newspapers of the ease of claiming refugee status in Canada"); Arbenz & 
ZiJrcher, The Co-crdxnation of the Asylum Policies of the European States, tn The Law of 
Asylum and Refugees: Present Tendencies and FirruRE Perspectives 126, 127 
(Council of Europe 1987) (reporting on travel agents taking asylum seekers to 



REFORMING ASYLUM ADJUDICATION 255 

that gravely threatens the overall structure of deliberate control over 
immigration — control that is also highly valued by the public, and by 
politicians and judges.®* 

Two public values, not often well articulated or conscious, but 
nonetheless strongly held, thus come into conflict in the asylum pro- 
gram.^ On the one hand stands the promise of refuge to the perse- 
cuted, on the other the demand for reasonable assurance of national 
control over the entry of aliens. If events force confrontation and a 
stark choice between the two, it seems likely as a matter of practical 
politics that the control principle would win over the refuge princi- 
ple.®' Refugee advocates should take this danger to heart since they 
have a major stake in minimizing the tension between the two goals 
by helping to structure a workable and reassuring asylum system. As 
a former UN Deputy High Commissioner, Richard Smyser, recently 
observed in a perceptive book: "The public will not aUow govem- 

Europe in the guise of tourists); Teitelbaum. Political Asylum in Theory and Practice, 76 
Pub. Int. 74, 79 (1984); Welame and Goodbye, Economist. Jan. 7, 1989, at S5; 
Applebome, Rings Smuggling U.S. Aliens Gain Sophistication, N.Y. Times, Mar. 2, 1989, 
at A23, col. 1. Most officials and a few private attorneys interviewed for this study 
also had tales about the schemes carried out by arrangers or organizers. 

65 See, e.g, Jean v. Nelson. 727 F.2d 957. 975 (11th Cir. 1984) (expressing 
concern that plaintiff's position "would ultimately result in our losing control over 
•our borders"), modified, 472 U.S. 846 (1985); T. Aleinikoff & D. Martin, supra note 
S, at 723 (quoting Attorney General Smith as stating that we, as a nation, had "lost 
control of our borders"); Public-Opinion Poll Reveals Support for Border Security, Immigr. 
Pol. & L., ^r. 20, 1989, at 5; see also Reitz. The Institutional Structure of Immigration as a 
Determinant of Inter-Racial Competition: A Comparison of Britain and Catuida, 22 Int'l 
Migration Rev. 117, ISl (1988) (discussing the importance of sustaining public 
perception of control over immigration). 

^ See E. Harwood. In Liberty's Shadow: Illegal Auens and Immigration 
Law Enforcement 10-15 (1986) (poll findings reflecting ambivalence on the part of 
Americans regarding immigration control); Pear. New Restrictions on Immigration Gain 
Public Support, Poll Shows, N.Y. Times. July 1. 1986. at Al, col. 5 (reporting polls 
indicating "that Americans have contradictory, ambivalent feelings about 
immigration"). Even more debilitating, each of the respective attitudes is likely to 
prevail at different points in the policy process: Restrictionism in the early stages of a 
perceived massive influx can be replaced by doubt and generosity as the time nears 
for placing identifiable individuals on a plane home. This is a recipe for stalemate. 
See Martin, supra note 1. at 12-13. 

6'^ In 1986, for example, the Swiss Parliament modified its law on asylum in 
response to the substantial increases in asylum applications that country was 
experiencing. In addition to enacting various procedural changes. Parliament made 
it easier for the Federal Council to derogate from the ordinary statutory guarantees 
of asylum. Theretofore such derogation was permissible only in times of armed 
conflict; since the 1986 law. derogation is also authorized in peacetime when there is 
an "extraordinary" influx of asylum seekers. See Loi sur I'asile du 5 octobre 1979. 
Recueil officiel des lois et ordonnances de la Confederation Suisse [RO] 1980 at 
1717. amended Modification du 20 juin 1986, art. 9(1). RO 1987 at 1674 (codified at 
Recueil systematique du droit federal [RS] 142.31). 



256 DAVID A. MARTIN 

ments to be generous if it believes they have lost control."^ Asylum 
will always be an inherently unruly component in an immigration 
system that usually functions with tidy categories and elaborate 
advance screening. But its unruliness can be curbed, and public sup- 
port thereby increased, if we can create a system capable of saying 
"no" to the unqualified — fairly, but firmly and expeditiously — while 
promptly welcoming the meritorious applicant.®^ 

B. Factual Issues 

In principle, it should be possible to distinguish between genu- 
ine refugees and those who do not qualify, thereby both honoring 
the asylum tradition and closing the loophole to those whose claims 
are meritless. But, for several important reasons, that task is far 
more difficult than it might initially appear. 

1. Lack of Clarity Concerning the Substantive Legal Standard 

Although Americans (along with most of the Western world) are 
virtually united in a commitment to protect refugees, they are far 
fi-om united in a common conception of "refugee." Everyday par- 
lance tends to treat anyone fleeing life-threatening conditions as a 
refugee, whether the source of the threat be natural disaster, foreign 
invasion, dvil unrest, or deliberate persecution.'^ The legal frame- 
work employs a narrower concept than this journalistic usage, and 
the 1951 Convention definition might be expected to provide the 
basis for a unified common understanding, built around the phrase 
"well-founded fear of persecution." But this phrase can also take on 
a variety of shapes, from highly expansive to narrowly crabbed, often 
depending, it seems, on whether the speaker wishes to include or 
exclude a particular group^ of claimants.' ' 

One common mistake, made by both the left and the right, is to 
assume that the existence of serious human rights problems in a 
country should translate into a finding that virtually all emigres from 



68 W. Smyser, supra note 20. at 119. 

^ See also The Select Commission on Immigration and Refugee Poucy, U.S. 
Immigration Poucy and the National Interest: Final Report 165 (1981) 
[hereinafter Select Commission] (reaching similar conclusions). 

'0 See generally Shacknove, Who is a Refugee?. 95 Ethics 274 (1985) (discussing 
the various conceptions of "refugee" held by society and selected international 
organizations). 

" S« C. Keely. Global Refugee Poucy: The Case for a Development- 
Oriented Strategy 6-1 1 (1981). 



REFORMING ASYLUM ADJUDICATION 257 

that country are refugees7^ Indeed, the language of the 1951 Con- 
vention can be made to fit this conception: If persecution occurs in 
the home country, any expatriate's claimed fear of it upon return is 
well-founded. The fear is not fanciful; it is based upon proven fact. 
If the legal standards conformed to this conception, adjudication 
would be gready simplified, for it could then be based on sweeping 
categorical judgments that would be easy to pronounce and 
administer. 

The case law makes clear, however, that the "well-founded fear" 
standard sets a higher threshold and ordinarily requires a far more 
individualized inquiry.'* Partisans in the debate over legal doctrine 
usually accept this narrowing gloss,'** even if it does not always pene- 



72 Such a reading underlies oft-heard complaints that our asylum policy is "out 
of sync" with our human rights policy, or more broadly, with our foreign policy. See, 
e.g., Dreifus. No Refugees Need Apply. Atlantic, Feb. 1987, at 32, 35 (quoting INS 
General Counsel Inman to this effect); Pear, supra note 40 (quoting an unnamed 
Reagan Administration official suting that deporting Nicaraguans was inconsistent 
with the lobbying effort to win support for the contras); Hansen, No Way to Treat 
Solidarity Refugeei, N.Y. Times. Apr. 1. 1985, at A21, col. 1 (arguing that INS denials 
of asylum to Poles undercut the credibility of our foreign policy). 

73 See. e.g., VUorio-Lopcz v. INS. 852 F.2d 1137, 1141 (9th Cir. 1988) (holding 
that there must be " 'specific evidence ... to support the alien's claim that 
persecution likely would be directed toward her as an individual' " (quoting Platero- 
Cortez V. INS, 804 F.2d 1127, 1130 (9th Cir. 1986))); Cardoza-Fonseca v. INS, 767 
F.2d 1448, 1453 (9di Cir. 1985), aff'd. 480 U.S. 421 (1987); Carvajal-Munoz v. INS, 
743 F.2d 562, 574 (7th Cir. 1984) (holding that an applicant must "present specific 
£acts establishing that she or she has actually been the victim of persecution or has 
some other good reason to fear that she or she will be singled out for 
persecution . . ."); In re Mogharrabi, Interim Dec. No. 3028, slip op. at 4 (BIA June 
12, 1987) ("It is clear that to a large degree the meaning of 'well-founded fear' can in 
&ct only be determined in the contexts of individual cases"). 

74 Some authors argue that protection of a wider range of endangered asylum 
seekers, such as those fleeing civil war. is now required as a matter of customary 
international law. See. e.g., Goodwin-Gill, supra note 53. at 902 (1986) (arguing that 
customary international law incorporates this mandate as a component of 
nonrefotdement); Perluss &: Hartman. Temporary Refuge: Emergence of a Customary Norm. 26 
Va. j. Int'l L. 551. 554-55 (1986) (arguing on the basis of an alleged "customary 
norm of temporary refuge"); see also Organization of African Unity Convention 
Governing the Specific Aspects of Refugee Problems in Africa. June 20, 1974, art. 1. 
1001 U.N.T.S. 46. 47 (This treaty's definition of "refugee" includes those covered by 
the definition of the 1951 Convention definition as well as those who flee "external 
aggression, occupation, foreign domination or events seriously disturbing public 
order"). But these authors generally do not dispute the conclusion stated in the text 
concerning the application of the 1951 Convention definition. See Hailbronner, Non- 
refoulement and "Humanitarian" Refugees: Customary International Law or Wishful Legal 
Thinktng?. 26 Va. J. Int'l L. 857. 880-87 (1986) (arguing that even sutes %vith 
expansive refugee policies have not surrendered the discretion to retreat to the 
narrower 1951 Convention standards). For a brief discussion of American practice in 
this regard, using the device of "extended voluntary departure," see st^rra note 18. 



258 DAVID A. MARTIN 

trate into public debate on the issue. Because legal standards 
demand individualized scrutiny, adjudication obviously requires a 
more sophisticated and difficult factual inquiry. The inquiry must 
first strive to assure the marshalling of all accessible information that 
might bear on the individual's circumstances and the conditions of 
the country. But just gathering that information is not sufficient, for 
in this contentious sphere, both governments and exiles may have 
significant reasons to distort their accounts of the facts or manufac- 
ture them from whole cloth. Crucially, asylum adjudication must 
therefore include the capacity to evaluate the assembled information 
in order to decide which is trustworthy and which is doubtful. Adju- 
dicators must have the authority, and the confidence, simply to reject 
some of the information tendered to them. 

The legal standards thus require that applicants show something 
more than simply that human rights abuses occur in the home coun- 
try.^* That "something more" is usually understood as a showing 
that the appUcant is likely to be targeted by the persecutors upon 
return. Even after the Supreme Court's Cardoza-Fonseca decision, 
there remains considerable room for dispute over just how much 
more of a showing this entails. Must the claimant show she would be 
"singled out" by the persecutors? How sharply focused must the 
threat be? What is relevant and probative evidence of such a threat? 
Must the applicant's testimony be corroborated? And further, what 
exacdy is "persecution"? Is sustained discrimination sufficient, or is 
something more such as a threat to life or freedom required ?^^ 

These disputes continue in the literature, in public debate, and 
in American and foreign case law. Although some progress has been 
made in refining the standards and achieving a more unified concep- 
tion, large differences of view abide. Ideally, the Board of Immigra- 
tion Appeals, the principal specialized custodian of legal doctrine in 
this field, would develop a body of doctrine that would refine and 
unify the understandings of the standard. The Board, however, has 
had difficulty playing this role, in part because it receives litde defer- 
ence from the reviewing courts in this field. Hence, splits among the 

75 See, e.g., Martinez Romero v. INS, 692 F.2d 595. 595-96 (9lh Cir. 1982) 
(holding that the reported anarchy in El Salvador, without other special 
circumstances, is insufficient to preclude return of an alien to that country by the 
INS). 

"^^ For a thorough exploration of the case law on these issues, see Blum, The 
Ninth Circuit and the Protection of Asylum Seekers Since the Passage of the Refugee Act of 1980, 
23 San Diego L. Rev. 327 (1986). 



REFORMING ASYLUM ADJUDICATION 259 

circuits develop and persist/' and the Supreme Court is not in a 
position to resolve more than a handful of such disputes. 

2. The Coast of Bohemia 

Compounding this substantive legal problem are the images we 
(both citizens and government officials) bring to judgments about 
asylum policy. The legal standard looks, in most cases, toward a 
finely calibrated individualized judgment of the risk of persecution 
the applicant would face in the homeland. The judgment must be 
based, to some extent, on general information about human rights 
conditions in the home country. The primary reliance, however, will 
fall, most of the time, on information specific to that individual.'® 

Public debate on asylum policy, however, proceeds in cruder 
terms. Partisans are often ready to make sweeping judgments, by 
nationality, about the merit of large groups of asylum seekers. Two 
leading schools of thought have been prominent in the debate. The 
first, which has long dominated actual outcomes, assumes that virtu- 
ally anyone from a Communist country would face persecution upon 
return. Holders of this view find it nearly unthinkable that the gov- 
ernment could contemplate deportation.'® A second school makes 

7*^ For example, there is currently a circuit split as to the meaning of persecution 
on account of a poUtical opinion. Compare Campos-Guardado v. INS, 809 F.2d 285, 
289 (5th Cir.) (requiring personally-held political opinion to qualify for asylum), cert 
denied. 484 U.S. 826 (1987) vrith Lazo-Majano v. INS, 813 F.2d 1432. 1435 (9th Cir. 
1987) (allowing personal reasons to qualify the applicant for asylum). There is also a 
split as to under what circumstances threats by guerrillas are sufficient to establish a 
"well-founded" fear. Compare Arteaga v. INS. 836 F.2d 1227. 1233 (9th Cir. 1988) 
(finding well-founded fear) with Rodriguez-Rivera v. INS. 848 F.2d 998. 1006 (9th 
Cir. 1988) (failing to find a well-founded fear). 

^B This statement must be qualified when the asylum claim is based on group 
characteristics, such as persecution directed at all members of a religious minority, or 
perhaps persecution that is so indiscriminate that virtually all who manage to escape 
should be recognized as refugees (I have in mind Pol Pot's Cambodia). For a helpful 
rethinking of definitional issues, see A. Zolberc. A. Suhrke. &: S. Acuayo. Escape 
From Violence: Confuct and the Refugee Crisis in the Developing World 25- 
33. 269-75 (1989). The authors break down the sociological concept of refugees into 
three subgroups: (1) activists, whose political actions draw the wrath of the 
government; (2) urgets, whose group affiliation (ethnicity, race, religion) is the 
reason for their oppression; and (3) victims, who are caught in a civil war's crossfire 
or are exptosed to generalized social violence. See id. Only the first two readily fit into 
the legal definition derived from the 1951 Convention. 

'9 See, e.g.. Pear, supra note 40 (reporting that rules being considered by the 
Reagan Administration "would establish a presumption that aliens fleeing 
[communist] countries had 'a well founded fear of persecution* and therefore met the 
statutory sundard for obtaining asylum in this country"); Hansen, supra note 72 
(suggesting that virtually all Polish nationals should qualify for asylum). 



260 DAVID A. MARTIN 

similar assumptions about Central American countries, particularly 
El Salvador and Guatemala.®^ 

a. The Essential Problem 

This kind of stereotyping or oversimplification is unfortunately 
commonplace — and to a significant extent inevitable — in public 
debate and policy decisions. In a classic work, Walter Lippmann 
explored comprehensively the influence on policy of these "pictures 
in our heads."®* In explaining how easily policymakers can err by 
relying on their own misconceptions about foreign lands, she wrote: 

[T]he real environment is altogether too big, too complex, and too 
fleeting for direct acquaintance. We are not equipped to deal with 
so much subtlety, so much variety, so many permutations and com- 
binations. And although we have to act in that environment, we 
have to reconstruct it on a simpler model before we can manage 
with it. To traverse the world men must have maps of the world. 
Their persistent difficulty is to secure maps on which their own 
need, or someone else's need, has not sketched in the coast of 
Bohemia.®^ 

The "coast of Bohemia" problem bedevils both public debate 
and adjudication in the asylum field.®^ But perhaps the image for 
our purposes should be shifted from the littoral to the physiographi- 
cal. Few nations enjoy a political geography characterized by a relia- 
bly fertile plain of steady himian rights observance. Outcroppings of 
abuses appear, sometimes intermittent hills, sometimes whole moun- 
tain ranges of severe persecution. The partisans in refugee 
debates — as well as adjudicators and judges under the current sys- 
tem — ^are too often inclined, in looking at nations to which they are 
favorably disposed, to mistake mountains for hills— or plains. The 
same people, in looking at nations to which they are hostile or for 

80 See, e.g.. Barker, New Wave of Salvadoran Immigrants Revives Call for Refiigee 
Status, Wash. Post, Feb. 18, 1989, at B3, col. 1. (indicating that the flood of refugees 
is caused by human rights violations). 

8' W. Lippmann, Pubuc Opinion 3 (1960). 

82 Id. at 16. For those who are rusty on their Eastern European geography, 
Bohemia is located in Western Czechoslovakia. It has no coast. 

8' The problem has been noted in a few cases, although not in these terms. See, 
e.g. Perez-Alvarez v. INS. 857 F.2d 23, 24 (1st Cir. 1988) (quoting from dissenting 
opinion of BIA Member Heilman, warning immigration judges against "assumptions 
regarding the way other societies operate," for they are often "proven to be totally 
wrong" upon a fiill hearing): Orantes-Hemandez v. Meese, 685 F. Supp. 1488, 1496- 
97 (CD. Cal. 1988) (finding that "impressions of INS agents and officials that 
Salvadorans come to the United States solely for economic gain" created barriers to 
the development of full and accurate information). 



REFORMING ASYLUM ADJUDICATION 261 

whose exiles they have (understandably) developed sympathy, often 
picture mountains where they should see hills, and then rush to the 
conclusion that the nation's exiles are refugees. Whatever the actual 
geography, it is also easy to forget that many people in those distant 
nations continue to inhabit the valleys even when the mountains 
loom large and forbidding. 

b. Boxes vs. Spectrums 

A related and persistent misunderstanding compounds the difR- 
culties in achieving a sensible and widely supported asylum policy, 
and also occasionally complicates adjudication. Much of the debate 
proceeds as though there are two sharply different categories of per- 
sons who find their way into the asylum adjudication system in this 
country: refugees, on the one hand, and economic migrants (or sim- 
ply "illegal aliens") on the other. A recent book on U.S. refugee 
policy (in other respects quite thorough and insightfid) reflects this 
attitude: 

Refugees are neither immigrants nor illegal migrants, 
although, like immigrants, they have forsaken their homelands for 
new countries and, like illegal migrants, they may enter those new 
countries without permission. But a refugee is, in the end, unlike 
either. Both the immigrant and the illegal migrant are drawn to a 
country. The refugee is not drawn but driven; often, he seeks not to 
better her life but to rebuild it, to regain some part of what she has 
lost.»^ 

Even if this sharply dichotomous view might, at one time, have 
captured the realities of refugee flows, it does not offer a helpful 
approach to today's asylum caseload. Today's dilemma is both tragic 
and surpassingly difficult precisely because, among current asylum 
applicants, refugees are so much like illegal migrants. Only an indis- 
tinct and difficult line separates those who should succeed on their 
asylum applications from those who should not. That is, most of 
those applying in the United States today were both drawn and 
driven, and they chose to come in response to a complex mix of 
political and economic considerations. Asylum seekers are not so 
different from the rest of us. We have a hard time deciding, particu- 
larly when we make difficult, life-altering decisions, and when we 
finally do choose a course of action, we act from a mix of motives. 

Take the case of a hypothetical Haitian farmer. For years she 

*♦ N.L. ZucKER & N.F. ZuCKER. mpra note 61, at xiv (emphasis added). 



262 DAVID A. MARTIN 

has been anguished by the unbridled power that local officials, sup- 
ported by militia units or bands of thugs, wield in the community. 
She knows of occasions on which they have terrorized those who 
resist their decrees, by burning a hut or killing a farm animal. But 
she also realizes that most people, who remain quiet, pay an occa- 
sional tribute, and raise no opposition, will be left alone. She 
endures this situation for many years, although she is more vocal 
than most of her neighbors about the community's problems. One 
year, her meager com crop fails owing to tlie worst drought of the 
decade. She worries how she will feed her children, already showing 
initisd signs of malnutrition. Shordy therezifter, she learns from 
friends about a boat that is about to depart for Miami, where, it is 
said, jobs are available and each day's pay exceeds a month's earn- 
ings in Haiti. They urge her to come along. 

The choice is not an easy one. She consults with her family, and 
they talk about the crop failure and the political miseries of the 
region. They discuss the pain that would come from lengthy separa- 
tion. They wonder whether the stories can be true about the jobs in 
Miami. Then they learn of a new episode of retribution visited on a 
farmer in a nearby valley, who apparendy ran afoul of the local hier- 
archy and was left maimed by a nightume attack. Some family mem- 
bers believe that this episode presages another serious outbreak of 
official violence; others think it is isolated and that things will quiet 
down as before. Pondering all of these factors, at first, she is sure 
she will stay. Then, after a child endures a week of persistent illness, 
she decides that she should leave, in an effort to earn enough to pay 
for medical care and food for her offspring. The family ultimately 
concurs, even though it means scraping together nearly all their sav- 
ings to meet the boat captain's fee for the journey. No thought is 
given to moving the whole family; they could not possibly afford the 
captain's charge. 

Once this person arrives in Miami, should she be seen as a 
"drawn" economic migrant or a "driven" refugee? She did not lack 
choice. Although many considerations strongly pointed toward leav- 
ing, she weighed a variety of difficult factors and chose at this time to 
travel to Miami. She could have stayed and remained subject to the 
same range of political and economic risks, uncertainties, and priva- 
tions. In fact, she was both drawn and driven, and the factors she 
considered were both economic and political.®* Does the presence 



8* Further evidence of the mix of motives appears, for example, in an extensive 
survey relating to Salvadorans in the United States. See S. Montes Mozo Se J. 



REFORMING ASYLUM ADJUDICATION 263 

of economic considerations undercut her refugee claim? Or should 
we try to assess which was her primary or dominant motivation for 
leaving?®® Or perhaps adjudication might center on the fact that the 
need to earn money for medical care — an economic consideration — 
was the immediately precipitating factor. 

Each of the perspectives implicidy underlying these questions is 
misleading. We do not need to find that she was only driven, nor 
assess what her primary motivation was, nor the immediately precipi- 
tating event. The best way to understand asylum adjudication is to 
focus on the degree of risk she would face when she returns. If the 
risk of persecution is sufficiendy substantial, her fear is well-founded, 
even if it was her need for funds to feed her children that sent her on 
the particular boat trip at the particular time. That she stayed home 
until economic considerations tipped the balance in her decision may 
be relevant — but only for the light it casts on the separate question 
concerning the degree of risk she truly faces. Her refugee claim is 
not forever tainted because she thought about jobs in Miami or the 
need for money to feed her family.®' Indeed, this would be abun- 
dantly clear if one could establish that shortly after her departure, 
which was precipitated, let us stipulate, by economic concerns, the 
local authorities expanded their violent suppression and actually 
began killing or jailing anyone who had ever publicly opposed the 
government or the local leaders. 

The 1951 Convention definition best translates into workable 
adjudicadve guidance only in this light. It does not ask how much of 
a role economics played in the decision to leave; it asks about risk 
levels upon return. The economic migrant/political refugee distinc- 
tion, however phrased, is misleading and unnecessary.®® 

Vasquez, Salvadoran Migration to the Unfted States: An Exploratory Study 
11-14 (1988) (28.5 percent of Salvadorans surveyed said they emigrated exclusively 
for political reasons; 20.6 percent claimed both political and economic reasons; 
apparendy the rest spoke only of economic or other non(>olitical reasons). 

8* Ste Anker je Posner, supra note 31, at 68 (suggesting— erroneously, in my 
view — that the applicant is a "refugee only if his 'primary motivation' is political"). 

8' See United States v. Santos- Vanegas, 878 F.2d 247, 252 (8th Cir. 1989). 

^ The majority opinion in Cardoia-Fonseca states that the definition "makes the 
eligibility determination turn to some extent on the subjective mental sute of the 
alien," and it later refers to "the obvious focus on the individual's subjective beliefs." 
INS V. Cardoza-Fonseca, 480 U.S. 421, 430-31 (1987). Odier commentary on the 
1951 Convention also claims equal status for "subjective" factors. See Office of the 
Unfted Nations High Commissioner for Refugees, Handbook on Procedures 
AND Crfferia for DETERMINING REFUGEE STATUS 11-12 (1979)[hereinafter UNHCR 
Handbook]. But no asylum adjudication system visited in the course of this study 
devotes any significant resources to inquiries into the applicant's subjective sute of 
mind; the fact of appUcation for asylum is taken as sufficient indication that the 



264 DAVID A. MARTIN 

If all asylum applicants did fit neady into one of two boxes — 
refugee or economic migrant — the adjudicative task would certainly 
be simplified. The job would simply be to unmask the impostors, 
those economic migrants who are base enough to pose as something 
they are not. Unfortunately, some people with authority over asylum 
decisions in Western countries sometimes speak of adjudications as 
though they did present such a mondity play. They hasten to label as 
abusive, fnvolous, or lawless those claims that simply fall short of the 
necessary showing.®® 

But the world is not that simple. Asylum adjudication, it must 
be recognized, is at best a crude and incomplete way to respond to 
the complex realities that the world presents.^ Our legal structure, 
for ultimately sound reasons, demands a simple yes or no answer to 
the asylum claim. But the dichotomous character of the results 
should not obscure the complexity onto which that yes-or-no g^d is 
forced. Asylum seekers present a spectrum of situations, with only 
subde shadings distinguishing the risk levels they face.®' Adjudica- 
tion must draw a line at some point on that spectrum. Furthermore, 
it must do so with care, so that it protects those whose risks exceed 
the threshold, even if they happen to have joined a mig^tion stream 

applicant holds a subjective fear of return. Moreover, the central holding of Cardoza- 
Fonseca clearly contemplates primary inquiry into the probability of |>ersecution — an 
objective determination. See Gibney, A "Well-Founded Fear" of Persecution, 10 Human 
Rts. Q. 109, 110 (1988). A British coun's effort to import a greater role for 
subjective fears (even if they could be shown objectively to be exaggerated) was 
overruled by the House of Lords in R. v. Sec'y of State for the Home Dept. ex parte 
Sivakumaran, 1 All E.R. 193 (H.L. 1988). 

89 For example, in a 1988 article, the then Minister of the Interior for the 
Federal Republic of Germany noted that only 9.9% of asylum seekers in 1987 were 
recognized as refugees. She then commented: "This implies that 90.7% of all those 
seeking asylum in West Germany unlawfully claimed to be |>oUtically persecuted." 
Zimmermann, View From West Germany, World Link. May 1988, at 65 (emphasis 
added); see, also INS Responds to Central American Influx, 65 Interpreter Releases 
1311-1312 (1988) (comments by INS Commissioner Nelson suggesting that 
unsuccessful asylum claims are "abusive"). A perceptive critique of such attitudes 
appears in Aleinikoff, Polttual Asylum in the Federal Republic of Germany and the Republic of 
France: Lessons for the United States, 17 U. MiCH.J.L. Ref. 183, 191-93 (1984). 

^ A pithy and revealing illustration of these complications, and of the effect of 
the pictures inside immigration judges' heads, appears in Neier, Cbsing Remarks, 16 
N.Y.U. Rev. L. & See. Change 157 (1987-88) (recounting experiences of the author, 
vice-chairman of Americas Watch, as an expert witness in asylum cases in which 
similar home-country consequences were seen as economic phenomena in Central 
America and political phenomena in Eastern Europe). 

91 See N.L. Zucker & N.F. Zucker, supra note 61, at 149 (recognizing this feature 
of much of refugee migration). 



REFORMING ASYLUM ADJUDICATION 265 

made up principally of those less severely threatened, who therefore 
lack, in this technical sense, a well-founded fear of persecution. 

c. Lessons 

These observations suggest two lessons with respect to asylum. 
The first focuses on the nature of public debate. Every effort should 
be made to avoid the use of dichotomous images and to break the 
ready links people rush to forge between human rights policy and 
asylum determinations. Obviously, an important relationship exists 
between human rights abuses in the home country and the merits of 
asylum claims by that country's nationals. But it is hardly a one-to- 
one correlation. Returning a high percentage of asylum seekers to a 
certain country is not necessarily inconsistent with a vigorous human 
rights diplomacy. Return pronounces no blessing on the home gov- 
ernment's overall practices; it simply indicates that these particular 
applicants did not make the requisite showing of the risk of persecu- 
tion. Similarly, granting asylum is not inconsistent with a policy of 
alliance and support for a democratically elected government. Many 
such regimes, particularly when newly installed, are only beginning a 
difficult process of curbing human rights abuses committed by the 
military, the police, or nongovernmental factions. We can support 
these efforts while still shielding the truly jeopardized targets of 
those incompletely controlled elements. 

Second, and more important for the immediate object of this 
study, the adjudication process must be shaped with attention to the 
"coast of Bohemia" problem. Asylum adjudicators are given an 
extremely difficult job, particularly in light of the inaccessibility of 
the facts they must develop, the potential consequences of their 
judgments, and the public ambivalence concerning their task. No 
wonder they may be tempted to retreat into categorical images about 
safety and danger in foreign countries that will make the sorting pro- 
cess easier.^^ Asylum reforms must therefore make allowance for 
this phenomenon and afford every opportunity, through initial train- 
ing, continuous supply of reliable information, and well-crafted mon- 
itoring, for a redrawing of the pictures inside the adjudicators' heads 
to conform more closely to the reality of political life in the home 
countries. 

92 See W. LiPPMANN, supra note 81, at 73-74. 



266 DAVID A. MARTIN 

3. Limited Accessibility of the Facts on Which 
to Base an Adjudication 

Little has been done to analyze carefully the various elements 
that go into the difficult determination of whether an asylum seeker 
has a well-founded fear of persecution, but such analysis is integral 
to designing an effective adjudication structure. In rough fashion, 
the determination may be broken down into three parts: (1) classical 
retrospective factfinding about past events specific to the claimant — 
adjudicative facts, in the terminology of Professor Kenneth Gulp 
Davis;®* (2) broader determinations about the practices of the gov- 
ernment or other alleged persecutors in the home country (often 
referred to as "country conditions") — legislative facts; and finally (3) 
an informed prediction (not truly a finding)** about the degree and 
type of danger the particular applicant is likely to face upon return. 
This last element is actually a prediction based on a combination of 
the first two elements. These are not three separate steps performed 
sequentially. They are closely interwoven, and it will appear that 
acquaintance with the second element is especially helpful in per- 
forming the first task effectively. Elach step presents unique chal- 
lenges, unlike those faced in other administrative adjudications. 
Indeed, of all such adjudications, asylum may rest on uniquely elu- 
sive factual foundations. 

a. Adjttdicative Facts 

To begin with, applicants typically base their claims on events in 
a distant land about which the U.S. government may have no 
independent information — ^matters such as their own past political 
activities, or specific abuses or threats directed at them or their fami- 
lies and friends. Of course, it is theoretically possible for the govern- 
ment to develop more information in an individual case, particularly 

9» &» 3 K. Davis. Administrative Law Treatise § 15.2. at 138-42 (2d ed. 
1980); Davis, An Approach to Problems of Evidence in the Administrative Process, 55 Harv. L. 
Rev. 364 (1942). This framework has been criticized, particularly because in some 
circumstances it may be difficult to tell whether specific matters constitute 
adjudicative or legislative facts. See, e.g., Robinson, The Making of Administrative Policy: 
Another Look at Rulemaking and Adjudication and Administrative Procedure Reform, 1 18 U. 
Pa. L. Rev. 485. 536-37 (1970). Nevertheless, the distinction is suflficiendy crisp to be 
quite illuminating in this setting. Judge Friendly used Davis's framework to make 
sense of the asylum adjudication process, and also to set forth limits on the 
appropriate use of State Department information in Zamora v. INS. 534 F.2d 1055. 
1062-63 (2d Cir. 1976). 

M See Washington Post Co. v. United Sutes I>ept. of Sute. 840 F.2d 26. 40-41 
(D.C. Cir. 1988) (Bork. J., dissenting). 



REFORMING ASYLUM ADJUDICATION 267 

once the applicant provides details, by assigning diplomatic person- 
nel posted to that country to investigate. But sheer expense pre- 
cludes such an effort except in a handful of cases, and the State 
Department freely admits that it rarely resorts to field checking. 
Such checking makes sense only where the information is likely to be 
reasonably accessible: for example, when it involves a well-known 
figure or relates to a large-scale event, such as a claimed political 
demonstration in the capital city, which can be readily confirmed or 
disproved. Moreover, even if the U.S. government would decide to 
dedicate greater resources to investigating more such cases, the 
investigations might yield little reliable information. If an ongoing 
threat of persecution truly exists, persons interviewed in the home 
country can hardly be expected to speak with candor to an unknown 
foreigner about such sensitive and dangerous matters. 

In short, even the straightforward retrospective factfinding 
involved in asylum determinations is difficult to accomplish. Bona 
fide applicants are unlikely to have left their homelands with cor- 
roborating documentation or with eyewitnesses to critical events. 
On the other hzmd, fraudulent applicants can probably count on the 
government's inability to produce evidence disproving their 
stories.®* 

Asylum determinations thus often depend critically on a deter- 
mination of the credibility of the applicant, for she will usually be the 
only available witness to the critical adjudicative facts of the case.®^ 
Because that person also has substantial incentives to lie or to 
embroider the truth (and few disincentives),®^ this makes for a sys- 



^^ At one time, administrators and courts tended to react to this problem with a 
dogmatic insistence on detailed corroboration of the applicant's claims. See Nasser 
V. INS. 744 F.2d 542. 544 (6th Cir. 1984); Shoaec v. INS. 704 F.2d 1079. 1084 (9th 
Cir. 1983); Kashani v. INS, 547 F.2d 376. 379-80 (7th Cir. 1977). Wisely, most U.S. 
authorities now recognize that the individual's own account may be the only available 
evidence regarding her own situation and that it should be accepted if reasonably 
detailed and consistent. See, e.g., Cardoza-Fonseca v. INS. 767 F.2d 1448, 1453 (9th 
Cir. 1985). aff'd. 480 U.S. 421 (1987); Ananeh-Firempong v. INS. 766 F.2d 621. 628 
(1st Cir. 1985); Carvajal-Munoz v. INS. 743 F.2d 562. 574 (7th Cir. 1984); In re 
Moifharrabi, Interim Dec. No. 3028, slip op. at 10 (BIA June 12, 1987) (stating that 
often the only evidence available is the applicant's own testimony, which is to be 
credited if it is "believable, consistent, and sufficiendy detailed to provide a plausible 
and coherent account of the basis for her fear"). 

^ See generally Watkins. Credibility Findings in Deportation Proceedings: "BearftngJ 
Witness Unto the Truth, 2 Geo. Immicr. L.J. 231 (1987) (the author is an immigration 
judge in San Diego). 

^^ In principle those who falsify their applications are subject to criminal 
prosecution under statutes punishing false statements. But the inaccessibility of the 
factual information obviously makes proof of knowing falsehood quite difficult, and 



268 ; DAVID A. MARTIN 

tern vulnerable to manipulation. I was struck, however, by the fre- 
quent comments from several participants I interviewed, particularly 
government decisionmakers and INS trial attorneys, indicating that 
the asylum system is saved from complete collapse largely by the 
admirable honesty of most of the applicants. 

Again and again in the course of the interviews carried out for 
this study, participants in the process, on all sides of the issue, 
emphasized one fact: In the vast majority of cases, the only useful 
detailed evidence respecting adjudicative facts comes from the 
mouth of the applicant. This feature must therefore figure promi- 
nently in any serious effort at procedural and structural reform. The 
system must be designed to mine as much useful information as pos- 
sible from the limited vein provided by the applicant's own testimony 
and her availability for questioning during any interview or 
hearing.^® 

b. Legislative Facts 

The second critical element of factfinding requires determina- 
tions about broader patterns of governmental behavior in the home 
country. For example, the asylum applicant may prove to the 
factfinder's satisfaction, through her own detailed testimony, that she 
was active as a union organizer for two years before leaving for the 
United States and that she heard stories of arrests of organizers in 
nearby towns before she left. But in order to assess the risk that the 
individual would face on return, the adjudicator must also learn from 
some source about relevant legislative facts. Does the government 
regard union organizers as opponents, subject to suppression? If so, 
what forms does the suppression take? Loss of a job or limitation of 
schooling options for organizers' children might not amount to per- 
secution (even though it would constitute a human rights violation), 
but beatings, jailings, or killings in reprisal for peaceful union activ- 
ity certainly would. If there have been some reports of such vio- 
lence, how widespread are the abuses? Were they based on the 
victim's union afRliation or on some other characteristic? In other 
words, is the current applicant relevandy similar to other persecution 

prosecutions are rarely, if ever, brought. The only sanction, then, is expulsion from 
the country following denial of the asylum claim — the same consequence faced if 
asylum had never been sought. 

^8 TTie BIA in essence recognized the importance of this process in a recent 
precedent decision that requires the asylum applicant to testify under oath at the 
hearing before the immigration judge; she cannot simply rest on the papers 
presented. See In re Fefe, Interim Dec. No. 3121, slip op. (BIA Aug. 1, 1989). 



REFORMING ASYLUM ADJUDICATION 269 

victims? And has there been a material change in the country since 
those events, such as a complete revamping of the police forces 
responsible for the earlier abuses, including reliable disciplining of 
the violators? 

Each of these questions will be difficult to answer, both because 
such patterns change over time, sometimes quite quickly, and 
because persecutors do not spell out the range of characteristics they 
seek in their victims. If the available information shows any substan- 
tial level of persecution of union activists, then uncertainties should 
be resolved in the individual's favor. Thus, the claimant should 
receive the "benefit of the doubt" commonly prescribed in works on 
refugee law.^ But the benefit of the doubt is hardly a magic 
formula, somehow dispensing with a need to reach a judgment about 
country conditions.*®® 

The individual applicant will not necessarily be in a position to 
provide insight on these wider matters, although the process should 
certainly allow for whatever assistance she or her counsel can pro- 
vide. Most Western countries therefore support their adjudicators 
with well-supplied documentation centers staffed by professionals in 
information science. Fortunately, the last twenty years have seen a 
welcome proliferation of human rights reporting, both by govern- 
ments'®* and private human rights organizations,'®^ as well as 

^ See UNHCR Handbook, st^mt note 88. at 48. 

^^ The UNHCR Handbook, for example, at one point states that adjudicators 
"are not required to pass judgment on conditions in the applicant's country of 
origin," although it does suggest that some such knowledge may help in the 
assessment of the applicant's credibility. Id. at 12-13. This is simply wrong. Such 
knowledge is useful, to be sure, in the latter setting. But the adjudicator cannot avoid 
passing judgment more broadly on country conditions as an unavoidable part of the 
ultimate decision on the merits. In the example in text, the applicant's credibility 
may be relevant only in deciding whether in faa she engaged in the union organizing 
activities she claims. The adjudicator might then proceed to recognize her as a 
refugee because she credits, wholly apart from anything the applicant says, numerous 
human rights reports describing a campaign of persecution directed against union 
leaders. 

•0* The most comprehensive is the annual series of human rights country 
reports prepared by the U.S. State Department. See, e.g.. Senate CoMMrrrEE on 
Foreign Relations and House CoMMnrEE on Foreign Affairs, 101st Cong., 1st 
Sess., Country Reports on Human Rights Practices for 1988 (Comm. Print 
1989). 

'02 In addition to comprehensive annual volumes published by Amnesty 
International covering most countries of the world, see, e.g.. Amnesty International. 
1988 Annual Report, many organizations publish topical reports on conditions in a 
single country or region as the need arises. See, e.g.. Amnesty International. 
Burma: Extrajudicdu. Execution and Torture of Members of Ethnic Minoiuties 
(1988); International League for Hin^tAN Rights and International Human 



270 DAVID A. MARTIN 

increasingly sophisticated efforts to systematize the information- 
gathering process and facihute sharing. '^'^ To minimize distortions 
in decisions about country conditions wrought by the needs of diplo- 
macy, many countries assure clear separation of adjudicators from 
their foreign ministries, so that diplomats become only one source of 
input.'*** Of course, such systems presuppose that adjudicators are 
equipped to sort through competing accounts and reach their own 
judgments about country conditions. 

Legislative facts should not be regarded, however, as simply 
something the adjudicator looks up or examines after she has com- 
pleted the proceedings addressed to finding the adjudicative facts, 
even though much of her knowledge about country conditions wiU 
doubdess come fi-om documentary sources rather than live testi- 
mony. Knowledge of political developments and patterns of perse- 
cution contributes toward making the final predictive judgment 
about risks faced if the individual returns home, but perhaps more 
importandy, such knowledge can also play a useful role in develop- 
ing and assessing the adjudicative facts themselves. 

This second use of knowledge about country conditions is often 
overlooked, but it remains crucial. An adjudicator thus equipped can 
better pick out those parts of the apphcant's story that are most rele- 
vant, and can ask specific questions that will flesh out the testimony 
in the most helpful fashion. '^^ Such expert questioning can also 
help expose inconsistencies and falsehoods more effectively. Since 
there are so few other checks on the asylum seeker's story (given that 
she is likely to be the only available witness to the key events), the 
system badly needs to make use of whatever other tools might be 
available for such assessment. But equipping adjudicators with such 

Rights Law GROin>, A Report on Human Rights in the Repubug of Korea 1980- 
1985 (1985); Lawyers CoMMmxE for Human Rights. El Salvador: Human Rights 
Dismissed— A Report on 16 Unresolved Cases (1986); Roth, Repression 
Disguised as Law: Human Rights in Poland (Lawyers Committee for Human Rights 
1987). 

'0' See, e.g., Thoolen, Refugees ond Information Technology, Refugees. Oct. 1988. at 
34. Nongovemmenul organizations have taken the lead in this sharing process with 
the active cooperation of the UNHCR, which has recently become the focal point for 
an international network. See Int'l Refugee Documentation Network, Circular No. 3 
(Mar. 1989) (reporting, inter alia, on recent efforts to strengthen the network, 
including training courses for documentalists and an internship program at 
UNHCR's Centre for Documentation on Refugees (CDR) in Geneva). 

'04 5** Aleinikoff. sttfrra note 89. at 234. 

'0* See Martin, Comparative Policies on Politual Asylum: Of Facts and Law, in 9 In 
Defense of the Auen 105. 109-1 1 (1987); Martin, The End ofDe Facto Asylum Toward 
a Humane and Realistic Response to Refugee Challenges, 18 Cal. W. Int'l L.J. 161. 167-69 
(1987-88) [hereinafter Martin, The Etui of De Facto Asylum]. 



REFORMING ASYLUM ADJUDICATION 271 

expertise is not just a device for spotting weaknesses or magnifying 
contradictions. Properly applied, it can also assist confused or inar- 
ticulate applicants in presenting fully the particularized information 
that will cast positive light on their claims. All this argues for making 
sure, to the maximum extent possible, that the adjudicators are 
themselves highly knowledgeable about country conditions. 

c. Predictive Judgment 

Finally, the information on the adjudicative and legislative facts 
must somehow be put together to reach a judgment as to the likely 
threat to the particular individual. For most of the countries from 
which current applicants come, it will be clear that persecution 
occurs at the hands of the government or societal elements beyond 
fidly effective control of the government. But what is the threat to 
this particular individual upon return? One must venture into the 
realm of prediction to decide. Cumulative expertise would also be of 
assistance here; such a judgment is not something that emerges rou- 
tinely fi"om the evidence placed on record in the case before a pas- 
sive adjudicator. 

It is also clear, however, that room for controversy will almost 
always remain. This is not a scientific prediction based on regular 
laws or formulas; it is an assessment that should be based, as much as 
possible, on conscientious attention to country condition informa- 
tion and individual fiaicts. As a result, the measure of an adjudication 
system's success cannot be the attainment of nearly universal accept- 
ance of the rightness of the results, particularly negative results lead- 
ing to deportation fi-om the country. Success consists instead in 
achieving sufficient acceptance of the process, including respect for 
the judgment and fairness of the decisionmakers, so that final grants 
and denials are regarded as authoritative. 

4. Difficulties of Cross-Cultural Communication 

One final complication deserves emphasis. As is apparent, much 
of the adjudication process will turn on assessment of the credibility 
of the applicant. Ordinarily, a decisionmaker judges credibility by 
probing the internal consistency and detail of the testimony about 
past events, observing the demeanor of the witness, and comparing 
the testimony to that person's earlier accounts or to other evidence 
regarding the events described. But because asylum applicants usu- 
ally come from cultures sharply different from that of the United 
States, these ordinary guideposcs to decision may not work well— or 



272 DAVID A. MARTIN 

at least they must be applied with considerable allowance for cross- 
cultural complications. '^^ These complications have been ably cata- 
logued and illustrated in a helpful article by Professor Kalin that 
should be read by all asylum adjudicators. '^^ Persons interviewed 
for this study, particularly UNHCR personnel and attorneys for asy- 
lum applicants, frequendy stressed that adjudicators must have the 
capacity to refrain from immediately applying tests and expectations 
derived solely from the culture of the haven state.***® 

Many asylum seekers come from societies where the population 
inherendy distrusts or fears government ofRcials (and often lawyers). 
Nothing in their past experience prompts them to open up readily to 
strangers, particularly when speaking of highly sensitive events. 
Thus, it is not surprising that in their first hours or even weeks or 
months in the United States they might fail to appreciate the new 
climate here that allows them to speak more freely and assertively. 
Many private attorneys interviewed for this study reported their own 
difficulties in winning trust and thus gaining candid accounts from 
their own clients. One experienced asylum attorney provided a 
graphic example. She reported that after spending nearly thirty 
hours with a reticent client, a Haitian asylum seeker, she believed she 
was prepared for a hearing. The day before the hearing, the client 
supplied some new information that revealed an entire new dimen- 
sion to the story that she had been afhiid to discuss earlier. Many 
more hours of patient interviewing were required to piece together 
the newly revezded true story, and concomitandy to bolster the cli- 
ent's trust in the attorney that would be needed for effective direct 
examination. 

Furthermore, psychological studies indicate that some of the 
strongest candidates for asylum may be those applicants with the 
greatest difficulties presenting their cases. Those who have been tor- 
tured or have witnessed the biiital slaying of friends or loved ones 
may have great difficulty retelling the key elements of their accounts. 



106 See generally Note, Credibility Findings in INS Asylum Adjudications: A Realistic 
Assessment, 23 Tex. Int'l L.J. 139. 142-43 (1988). 

1^'^ See Kalin, Troubled Communication: Cross-Cultural Misunderstandings in the Asylum- 
Hearing, 20 Int'l Migration Rev. 230 (1986). 

'08 A UNHCR officer in Canada suggested that the most important quality to be 
sought in an adjudicator is empathy: "Can this person understand what real refugees 
go through?" She emphasized, however, that people with this quality "can still be 
firm in saying no" when that is warranted; fkilure to reject the unqualified, she 
stressed, also "screws up the system." Interview %vith Paul LaRose Edwards, in 
Ottawa. Canada (Dec. 20, 1988). 



REFORMING ASYLUM ADJUDICATION 273 

At times, post-traumatic stress disorder may even block consistent 
memory of past events. *°^ 

American decisionmakers unaware of these complications are 
likely to seize upon inconsistency and reticence (particulariy about 
matters that most Americans would regard as of the g^reatest impor- 
tance) as evidence of dissembling — ^for reasons that usually hold 
good within our own cultural context. These actions are not such 
ready signals of dissembling, however, when the individuals involved 
come from sharply different cultural backgrounds. A reformed sys- 
tem must equip its decisionmakers to avoid snap judgments and 
make adequate aUowance for cross-cultural difficulties. But such 
awareness must also be employed with care. Some of the writings 
about cross-cultural sensitivity seem almost to assume that there is 
an innocent explanation for any stumble, vagueness, or change of 
story.**® This is not necessarily the case. A reformed system must 
also equip its adjudicators to sort through such phenomena and still 
be able to spot false tales — because sometimes inconsistency and ret- 
icence really do result from falsehood and not from more innocent 
explanations. The line to be walked is a fine one. 

C. The Imperative of Speedy Final Decisions 

The foregoing problems suggest genuine, indeed profound, dif- 
ficulties in applying the legal standards with precision and fidelity. 
As long as these problems impair accurate decisionmaking, the sys- 
tem will have its own built-in magnet effect. When the process can- 
not reliably sort the qualified from the unqualified, asylum applicants 
drawn to the system will include not only those with a reasonable 
chance of qualifying, but also others whose claims are marginal or 
nonexistent. They will come hoping to take advantage of these very 
weaknesses to gain an undeserved benefit, namely, the award of asy- 
lum status and possibly eventual permanent residence. 

In later sections, I propose measures to make the maximum use 



'(^ See, e.g., Allodi, Randall, Lutz. Quiroga, Zunzunegui, KolfT, Deutsch & Doan, 
Physical and Psychiatric Effects of Torture: Two Medical Studies, in The Breaking of Bodies 
AND Minds: Torture, Psychiatric Abuse, and the Health Professions 58, 65-72 
(E. Stover &: E. Nightingale eds. 1985); Note, supra note 106, at 148-50; Rovner, The 
Torture of the Refugee: Why Judges Don't Believe, Wash. Post. Sept. 2. 1986 (Health 
Section), at 10, 11. 

1 >o For a court decision that verges on this approach, rejecting the immigration 
judge's negative credibility finding and asserting that earher untrue statements and 
changed stories actually supported the persecution claim, see Turcios v. INS. 821 F.2d 
1396. 1400 (9th Cir. 1987). 



274 DAVID A. MARTIN 

of the available information sources in service of the goal of accu- 
racy. One might think that such an achievement would suffice to 
accomplish the fundamental objectives of our asylum program and 
also curb the magnet effect — by providing asylum to the persecuted 
and saying "no" to those who seek to use asylum mainly as a looj>- 
hole. If accuracy were all we had to accomplish, we could embrace 
elaborate schemes that promise to serve that goal, even if they con- 
sumed a large amount of time to reach final decisions and employed 
multiple layers of administrative and judicial consideration as a check 
against error. Unfortunately, however, the magnet problem is more 
complex, in ways that demand an end to undue layering so as to pre- 
serve the capacity for speedy final and enforceable determinations. 

1 . The Scope of the Magnet Elffect 

The magnet effect is not solely the product of perceived chances 
to gain full asylum despite a weak case. Benefits that applicants can 
expect to enjoy before a final ruling is issued in the case — a period that 
now can stretch for months, and usually lasts years — also contribute 
importandy to that phenomenon. Of course, to some extent both 
accuracy and speed are goals of any administrative adjudication sys- 
tem. But the need for expeditious finality is more intense here. In 
other adjudication processes, such as those governing disability 
claims or public welfare or licensing, the applicant ordinarily does 
not enjoy the benefit sought until there has been a determination on 
the merits that she fidly qualifies. Nothing in the application and 
waiting process itself tempts the unqualified to clog the system. With 
politiccd asylum, in contrast, the simple act of applying has usually 
brought important benefits that magnify the attractions, whatever 
the ultimate determination on the merits. With a few recent excep- 
tions, the very act of applying for asylum has resulted, after a brief 
delay, in the issuance of preliminary papers that both authorize 
employment and permit free movement within U.S. territory. TTiese 
two features comprise the bulk of the main benefits expected firom 
asylum itself, particularly for those who know they have at best weak 
claims."' 

The longer the period of enjoyment that comes simply from the 

HI To be sure, the uncertainty about the duration of such benefits is a 
disadvanuge. But it is a disadvantage that is doubtless felt more acutely by true 
refugees for whom there are genuinely grave risks if they return. Those who know 
they face little risk at home harbor fewer concerns about what will transpire once the 
asylum-seeker suge ends. In the meantime, they have nearly complete access to the 
job market and the other features of life in a suble, wealthy, and free society. 



REFORMING ASYLUM ADJUDICATION 275 

act of applying for asylum, obviously the greater the attraction in fil- 
ing a marginal claim. When word gets back to the home countries of 
those who initially benefit from such arrangements during the asy- 
lum-application stage, more and more people with marginal or non- 
existent claims are likely to come, hoping to achieve at least the 
benefits of years of productive working life in a wealthy country, 
whatever the ultimate outcome.**^ A successful asylum system must 
thus place a high priority on speed in adjudications (including all 
stages of review) in order to avoid these incentives for marginal asy- 
lum seekers. 

2. The Alternative of Deterrent Measures 

Of course, speedier final decisions are not the only way to elimi- 
nate the artificial attractions of the asylum-seeker stage. One could 
simply end the provision of free movement and work authorization 
during this period. Many Western countries have been moving in 
this direction, imposing a variety of restrictions and deterrents that 
have raised the concern of UNHCR and provoked harsh condemna- 
tion from the nongovernmental organizations (NGOs) that support 
asylum seekers and advocate refugee causes. These restrictive prac- 
tices include denials of work authorization, enforced housing in aus- 
tere communal facilities, other limits on fi-eedom of movement, and 
sometimes fidl-scale detention in jail-like fecilities."' 

"2 Canada, for example, experienced an exponential growth in asylum claims 
filed by nationals of Trinidad and Tobago throughout 1987-88 (reaching 2739 such 
applications in 1988). Once its new system was implemented on January 1. 1989. 
promising swift rejection of manifesdy unfounded claims, applications from Trinidad 
dropped to 15 in the first quarter of the year. See 63 DocuMENTATiON-RiFUCits 1, 6 
(Jan. 25-Feb. 3. 1989); Refugee Determination in Canada: First Quarter Review 9 
(Apr. 25, 1989). For a more complete description of the new Canadian adjudication 
system, which appears to hold considerable promise, see the expanded version of this 
study (forthcoming Yale University Press). See also Blum 8c Laurence, Cold Winds from 
the North: An Analysis of Recent Shifts in North American Refugee Policy, 16 N.Y.U. Rev. L. & 
See. Chance 55 (1987-88); Hathaway. Postscript. 34 McGiix L.J. 354 (1989); 
Hathaway, Selective Concern: An Overview of Refugee Law in Canada, 33 McGill L.J. 676 
(1988); Rusu, The Development of Canada's Immigration and Refugee Board Documentation 
Center, 1 Int'lJ. Refugee L. 319 (1989). 

"5 See generally Fullerton. Restricting the Flow of Asylum-Seekers in Belgium, Denmark, 
the Federal Republic of Germany, and the Netherlands: New Challenges to the Geneva 
Convention Relating to the Status of Refugees and the European Convention on Human Rights, 
29 Va. J. Int'l L. 35 (1988); Martin, supra note 1. The United States has also 
employed Coast Guard interdiction of vessels coming from Haiti. Although the 
interdiction process is designed to include screening to permit persons with valid 
refugee claims to come to the United States, interdiction has .voked severe criticism. 
See T. Aleinikoff & D. Martin, si^a note 3, at 724-26; Helton, Political Asylum under 
the 1980 Refugee Act: An UnfulfiUed Promise, 17 U. Mich. J.L. Ref. 243. 254-56 (1984). 



276 DAVID A. MARTIN 

Considerable misunderstanding has arisen regarding the use of 
deterrent measures and restrictive practices. NGOs sometimes act as 
though any deterrent steps are illegitimate — sheer vindictiveness vis- 
ited upon innocents, many of whom may prove to be bona fide refu- 
gees. Some perspective is needed. Designing policy to discourage 
the unqualified from even applying for a benefit is a perfecdy legiti- 
mate policy objective, particularly when existing statistics demon- 
strate that a high percentage of apphcations lack merit.*** To the 
extent that current measures are meant to encourage self-selection, 
so that only those with strong cases bother to leave their home coun- 
tries, they address an unimpeachable administrative aim. In design, 
at least, these restrictive practices are meant to send a "genend 
deterrence" message to persons still in the home country. 

The message grows more complex in practice. At this prelimi- 
nary stage, deterrent measures almost inevitably apply to all asylum 
applicants, whatever the streng^ or weakness of their claims. (To 
sort the strong from the weak at this stage — save for screening out 
wholly frivolous applications — ^would simply be too cumbersome, 
because of all the factual difficulties canvassed in the previous sec- 
tion.) Thus, the burdens of these measures often fall on genuine 
refugees.* ** It is not wholly surprising, then, that some judges might 
view these measures as penalties for filing an asylum claim, or as 
coercion meant to force current applicants to withdraw their applica- 
tions and return home, rather than as deterrents addressed to those 
still in the home country, designed to convince them (if they are not 
substantially threatened) to decide against coming to the United 
States. If they see such measures this way, courts are likely to declare 
the deterrents invalid for conflicting with the statutory right to apply 
for asylum.**® 



1 14 Naturally, a major part of the NGO criticism stems from a belief that existing 
grant rates are woefully inadequate. But imless the grant rate should approach 100 
percent, encouraging self-selection remains a legitimate aim. 

••* Some interviewed for this study, however, who advocated the detention of 
asylum seekers, argued (with some measure of plausibility) that the true refugee 
should be the least affected by such deterrents. They should see humane detention 
in the United States, for a limited period that will definitely end with a final 
adjudication, as a small price to pay to escape from actual persecution in the 
homeland. True refugees, such persons argued, would not begin to think about 
withdrawing the application and returning home, for the fate there is clearly worse. 
Only those facing httle real risk at home might be tempted ("coerced," as the other 
camp would describe it) to withdraw an application in order to end the detention. If 
the fate at home is that much less intimidating than a few months in detention here, 
detention advocates argue, then such persons do not merit asylum in any event.' 

"6 See. e.g.. Orantes-Hemandez v. Meese, 685 F. Supp. 1488, 1510 (CD. Cal. 



REFORMING ASYLUM ADJUDICATION 277 

The basic problem is that these deterrent measures and restric- 
tive practices are indiscriminate in their impact. By their very nature, 
they fall equally on deserving refugees and the most flagrant abusers 
during the asylum application stage. A case could even be made that 
they fall with more debilitating effect on true refugees, because 
leng^y uncertainty over their ultimate fate, coupled with enforced 
idleness and perhaps prison-like detention, will carry the most severe 
psychological impact for those who know with substantial assurance 
that persecution awaits them at home. (It may be even more devas- 
tating for those who have already been tortured or severely mis- 
treated.) For these reasons, such deterrents plainly should be 
avoided if workable alternatives are available.* " At best these deter- 
rent measures are crude tools, meant to send a message to marginal 
applicants discouraging them from leaving the home country, but 
capable of implementation only by imposing harshness on true refu- 
gees as well, those who will ultimately be found to merit asylum. 

3. Toward a Better-Targeted Deterrent 

We need instead a discriminate deterrent, more precisely 
focused on the marginal cases, and one that takes away the artificial 
attractions of the asylum applicant stage of the proceedings. Such a 
deterrent is available, if we can change the adjudication system to 
achieve one crucial result: the prompt reappearance in the home vil- 
lage of applicants whose cases were at best marginal. Such an event 
makes apparent to others similarly situated that such a trip is not 
worthwhile; they will not be able to work long enough even to repay 
the "travel agent's" fee.**® Speedy finality is the essential precondi- 

1988) (granting injunction because INS practices and procedures at detention 
centers "discourage class members from exercising and pursuing asylum"); Nunez v. 
Boldin. 5S7 F. Supp. 578, 586 (S.D. Tex.) ("The present procedures do not 
sufficiently assure that genuine asylum claims will be heard . . . ."), appeal dismissed, 
692 F.2d 755 (5ih Cir. 1982). Other courts, however, have been more willing to 
sustain detention schemes, including their use for explicit deterrent purposes. See, 
e.g.. Amanullah v. Nebon, 811 F.2d 1, 15 (1st Cir. 1987); Singh v. Nelson, 623 F. 
Supp. 545. 556-61 (S.D.N.Y. 1985). See generally Helton, The Legality of DeUuning 
R^ees m the United States. 14 N.Y.U. Rev. L. &: Sec. Chance 353 (1986); Schmidt, 
Detention of Aliens, 24 San Diego L. Rev. 305. 329-33 (1987) (discussing detention of 
asylum applicants). 

"7 For an elaboration of these points, see Martin, supra note 1. 

"B Some critics of an earlier version of this study made much of the lack of 
proof that such returns would really deter other migrants. It is true that no empirical 
studies support this claim. Nor are there any, to my knowledge, that disprove it. 
Reliable testing procedures would be difficult to develop. 

Suggestive support can be found, however, in much recent social science 



278 DAVID A. MARTIN 

tion of this deterrent. The message is lost if two or three years pass 
between depanure and return, particularly if the apphcant has been 
working while the amplication was pending. 

How much speed is necessary? We lack empirical data to calcu- 
late the outside limits with any kind of accuracy (and anyway the cal- 
culations would vary by country and by travel agent). But if all but 
the most complicated cases could reach finality within six to nine 
months, including all the stages of consideration and review, htde in 
the application process would add artificially to the attractions of the 
asylum system. 

NGOs that strongly oppose indiscriminate deterrent practices — 
for good and worthy reasons — should remember that the best way to 
help avoid them, or reduce their harmful impact, is to cooperate with 
the government in fashioning a speedy and accurate system that can 
accomplish discriminate deterrence. This point cannot be overempha- 
sized. In the absence of the capacity to make final decisions quickly, 
officisds have no way to respond to legitimate public concerns over 
massive influxes, unless they turn to the deterrent measures and 
restrictive devices that NGOs and the UNHCR condemn. To defeat 
prudent streamlining of the adjudication process is to invite reliance 
on cruder measures of deterrence. Mere nominal acceptance of the 
need for expeditious proceedings is not enough. Refugee advocates 



literature. Criticizing simplistic "push-pull" models of migration, these woriu 
emphasize the role of social networks in encouraging and sustaining migration. That 
is, if the first emig^rants from a particular community succeed in establishing 
themselves in a new country, their experience, communicated homeward (often 
along with significant remittances), encourages others from the same locale to make 
the trip. Moreover, their presence in the target city or town within the new country 
helps others during the difficult early months. See, e.g., Boyd, Family and Personal 
Networks in Intematumal Migration: Recent Developments and New Agendas, 23 Int'l 
Migration Rev. 638 (1989); Massey, Understanding Mexican Migration to the United 
States. 92 Am. J. Sec. 2 (1987); Portes & Borocz, Conlemporary Immigration: Theoretical 
Perspectives on its Determinants and Modes of Incorporation, 23 Int'l Migration Rev. 606 
(1989). 

Although such studies generally address "guest workers" and similar economic 
mig^tion and do not focus on asylum-seeker networks, there is no reason to believe 
that this phenomenon would fail to operate in the latter context. Much anecdotal 
evidence about the role of asylum "travel agents" and similar entrepreneurs fits 
readily within these network theories. See supra note 64 and accompanying text. 
Moreover, the disparate caseload patterns in adjoining asylum countries also suggest 
a network effect. In 1989, for example, Canada received 1,966 Sri Lankan asylum 
seekers, see [Canadian] Immigration and Refugee Board, News Release of Jan. 19, 
1990, at 4; the United States only 40. See INS, Asylum Cases Filed wtth District 
Directors. Fiscal Year 1989 through Sept. 1989 (CORAP ubie). Swift 
deporution of unsuccessful asylum seekers would seek to break that chain and would 
be designed to send a very different message back to the source communities. 



REFORMING ASYLUM ADJUDICATION 279 

will have to join in making difficult decisions about the trimming of 
certain procedures (which will undeniably carry some costs to the 
goal of accuracy) in order to achieve truly speedy determinations. 

A second point concerning restrictive measures and deterrents 
may be more immediately relevant. From the government's stand- 
point, most of these other restrictions do not eliminate the priority 
for speed; they simply create other reasons for embracing it as a vital 
goal. The U.S. government probably cannot simply deny work 
authorizations (now seen by some as a major contributor to an artifi- 
cial magnet effect) to asylum applicants without establishing some 
other public scheme to provide for the subsistence needs of the idled 
asylum seekers until they are either recognized as refugees or 
removed fi-om the country. Whether such provision is made in com- 
munal facilities or in actual detention centers, it will still require a 
substantial commitment of public resources. Every day of added 
delay, therefore, compounds the expense imposed on the public 
treasury.**® 

There is a final reason for embracing speedy procedures, 
derived fi'om the perspective of the legitimate and meritorious claim- 
ant — for whose benefit, after all, asylum protections were initially 
adopted. Initial decisions in many district offices now take six to 
eight months, largely because of backlogs created by the overload of 
asylum applications. Bona fide applicants with qualifying cases 
should not have to wait so long to have the burden of uncertainty 
lifted from their shoulders. As indicated, their primary need is to 
find the calm and security that will enable them to rebuild some sem- 
blance of a normal life. They are much more likely to make a suc- 
cessful transition (including recovering fi'om past episodes of torture 
or other traumatic mistreatment) if security, in the form of a settled 
and reasonably permanent immigration status comes quickly after 
arrival. 



^^^ See Appendix (for cost estimates). In faa, truly expeditious and accurate 
procedures might further support the use of detention or enforced housing 
arrangements and denial of work authorization during the asylum-seeker stage. The 
main objection to these measures has been their baleful impact on bona fide 
claimants. But if most bona fide claimants can be recogpnized in the first-round 
proceeding and thus need wait only a short period (say two to three months) in such 
a setting, much of the unintended coercive impact disappears. The Select 
Commission on Immigration and Refugee Policy in fact recommended such 
arrangements, focusing on accommodation in "federal asylum processing centers" in 
its recommendations for coping with "asylum emergencies." These 
recommendations appear to have presup|x>sed speedy determinations. See Selxct 
Commission, sufira note 69. at 165-68. 



280 DAVID A. MARTIN 

Speedy finality is imperative. It must not be achieved at the 
complete expense of either accuracy in outcomes or fairness in the 
process, but some tradeoffs will be necessary. Speed here is not sim- 
ply the kind of virtue it may be in some other administrative set- 
tings^-desirable but optional, a pleasing accomplishment if 
achievable, but not gravely damaging if other aims preclude its 
attainment. Speedy denials of unworthy asylum applications, fol- 
lowed by prompt deportation, are indispensable if we are to imple- 
ment the only really humane deterrent available to the system. In 
time of large-scale influx, the inability to deport the unqualified in 
fairly short order will lead governments to resort to other cosdy and 
troublesome deterrents which indiscriminately burden genuine 
refugees. 

III. The American Adjudication System 
A. Historical Background 

The earliest American regulations estabhshing procedures for 
asylum and related adjudications appeared in 1953, implementing 
the 1952 version of § 243(h) of the Immigration and Nationality Act 
(INA).*^*' They provided for "interrogation under oath by an immi- 
gration officer" to examine the claim that the alien would be subject 
to physical persecution if deported. The regulations permitted the 
presence of an attorney, at the alien's expense, but said nothing 
more about the attorney's role during the interrogation. Final deci- 
sions, presumably based on the record of the interview, were to be 
rendered by the Commissioner or Assistant Commissioner — a cum- 
bersome requirement changed a year later to vest that authority 
instead in regional commissioners of the Immigration and Naturali- 
zation Service (INS).*^' 

In 1962, new regulations took effect establishing different 
arrangements for persecution claims in deportation proceedings. 
They established specific procedures for the immigration judge (then 
called a special inquiry officer) to designate an alternate country of 
deportation, in case the country the alien chose refiised to accept 
him, and they required the judge then to advise of the possibility, 
under § 243(h) (which was then still a discretionary provision), of 
withholding deportation to that country. If the alien chose to claim 
that protection, she received ten days to file an application docu- 

•20 See 18 Fed. Reg. 4925 (1953) (adding 8 C.F.R. § 24S.S(b)(2)). 

•2> See 19 Fed. Reg. 9172. 9179 (1954) (amending 8 C.F.R. S 243.3(b)(2)). 



I 



REFORMING ASYLUM ADJUDICATION 281 

mendng her persecution claims, after which she would be examined 
under oath on these issues as in a deportation proceeding. '^^ Later 
amendments made clear that an INS trial attorney could also intro- 
duce evidence bearing on the persecution claim. *^* 

By 1962, then, the two basic patterns for asylum and related 
adjudications that our system has known had already emerged. On 
the one hand, INS made some determinations based on a nonadver- 
sarial interview or interrogation carried out by an immigration 
officer. In other settings, immigration judges decided whether to 
provide relief from deportation after more formalized trial-type pro- 
ceedings. In succeeding years, INS refined the nonadversarial proce- 
dures, vesting the ultimate decisionmaking authority not in the 
regional commissioners or higher officials, but in the district direc- 
tors, who of course relied in practice, most of the time, on the immi- 
gration officer who conducted the interview. As INS discovered 
more and more settings (outside the deportation procedure) in 
which persecution claims might arise, it adopted a variety of new reg- 
ulations and instructions specifying that district directors, rather 
than immigration judges, were to hear and determine those matters. 
For example, district directors received authority to make final deci- 
sions on persecution claims by alien crewmen,*^** excludable 
aliens,*^^ and applicants for the special benefits of INA § 203(a)(7) 
who were already in the United States.*^® 



122 See 26 Fed. Reg. 12.110. 12,112 (1%1). With a few modificauons. this 
provision, 8 C.F.R. S 242.17, remains in the regubtions and affords one avenue for 
consideration of withholding by the immigration court. Most applicants today, 
however, affirmatively apply for asylum much earlier in the immigration court 
proceedings, under other regulations. 

»23 See 39 Fed. Reg. 25.642 (1974) (amending 8 C.F.R. § 242.17(c)). 

•s-* See 32 Fed. Reg. 4341. 4342 (1967) (amending 8 C.F.R. § 25S.l(f)). This 
assignment of authority to the district directors, to the exclusion of special inquiry 
officers (now called immigration judges), was approved, over vigorous dissent, in INS 
V. Stanisic, 395 U.S. 62 (1969). Interestingly, this provision, which provides 
protections only to those fearing persecution in Communist countries, was not 
amended in the wake of the Refugee Act of 1980, which mandated neutral and 
apolitical standards in refugee matters. It survives as a kind of dinosaur in the Code 
of Federal Regulations. Crewmen from other countries, however, may still claim 
asylum under other provisions. 

•25 This was initially established in the less formal "Operations Instructions," 
and enshrined in the regulations only in 1974. See Pierre v. United States, 547 F.2d 
1281, 1285 (5th Cir.), vacaUd and remanded to consider mootness, 434 U.S. 962 (1977). 

•26 See 30 Fed. Reg. 14,772, 14,779 (1965) (adding 8 C.F.R. § 245.4). ammded, 
34 Fed. Reg. 19.799 (1969). INA § 203(a)(7) was enacted primarily to authorize 
overseas refugee programs for the resettlement of persons who fled persecution in 
Communist countries or countries "within the general area of the Middle East." See 
8 U.S.C. § 1153(a)(7) (1976). But a proviso allowed for adjustment of status to 



282 DAVID A. MARTIN 

In 1970, a Lithuanian seaman named Simas Kudirka was forcibly 
returned to a Russian vessel a few hours after she had escaped to a 
U.S. Coast Guard cutter. Although INS had not been involved in 
this extraordinary incident, the ensuing outcry launched a complete 
review of asylum procedures used by all agencies. *^^ Eventually INS 
promulgated new asylum regulations, permitting both excludable 
and deportable aliens to apply to the district director for asylum on a 
new form. Form 1-589. TTiese 1974 regulations also made specific 
provision for a practice that had already taken root. They required 
the district director to seek the views of the State Department on an 
asylum claim, while also giving the alien an opportunity to explain or 
rebut any State Department comments before a decision could be 
based thereon. Such comments were not binding on the adjudicator, 
but if the district director decided to deny an application despite a 
favorable State Department letter, she had to certify her ruling to the 
regional commissioner for final decision. *^^ 

Under the 1974 regulations, any deportable alien denied asylum 
could, in essence, renew the claim in deportation proceedings by 
applying to the immigration judge for protection under § 243(h). 
However, for excludable aliens, those apprehended at the border 
before msdung an entry into the United States, the district office 
remained the only venue for an asylum claim. Within a few years, 
excludable Haitians challenged, on due process grounds, the regula- 
tions' failure to permit an "evidentiary hearing" (of the type pro- 
vided in immigration court proceedings) when so much was 
potentially at stake. *^ District courts split on the issue, but the Fifth 
Circuit eventually ruled for the government, approving the regula- 
tions. *'° While the asylum seekers* certiorari petition was pending 
in the Supreme Court, however, the newly installed Carter Adminis- 



pcrmanent resident of those who had been physically present in the United States for 
two years and could prove that they qualified. See generally T. Aleinikoff ie D. 
KfARTiN. supra note 9, at 622-2S (discussing this version of INA fi 203(a)(7)). 

127 The Sute Department's resulting policy was announced in 1972 and 
remains in force today with slight changes. See Public Notice 351. Request for 
Asylum— Policy and Procedures. 37 Fed. Reg. 3447 (1972). modified. Public Notice 
728. Requests for Asylum. 45 Fed. Reg. 70.621 (1980). 

>28 See 39 Fed. Reg. 41.832 (1974) (adding 8 C.F.R. pt. 108). amended, 40 Fed. 
Reg. 3407. 3407-08 (1975). 

1^ This had long been a matter in contention; it was the central dispute 
between the majority and dissent in a Supreme Court case construing the regulations 
governing asyliun claims filed by alien crewmen. See INS v. Sunisic, 395 U.S. 62 
(1969). 

•so See Pierre v. United Sutes. 547 F.2d 1281, 1290 (5th Cir.), vacaud and 
remanded to consider mootness. 434 U.S. 962 (1977). 



REFORMING ASYLUM ADJUDICATION 283 

tration decided to accede to the plaintiffs' demands, finding merit in 
the due process concerns and seeing no major costs if the change 
were made.'*' 

In considering new regulations to implement that 1977 conces- 
sion, INS attempted to take to heart the vigorous objections to dis- 
trict director adjudications that had been voiced throughout the 
earlier litigation — assertions which made it sound as though such a 
setting could never provide justice in asylum cases. '^^ In conse- 
quence, the agency promulgated regulations in 1978 that would have 
made the immigration court, with very limited exceptions, the only 
forum for consideration of such asylum claims, for either excludable 
or deportable aliens.*'* 

To INS's legitimate surprise, refugee advocates suddenly dis- 
played a change in perspective. They filed comments on the regula- 
tions, as well as briefs in litigation, that revealed a remarkable 
rediscovery of the virtues of nonadversarial proceedings before the 
district directors, exactly the officials whose decisions had been so 
heavily criticized in the course of the earlier litigation. Nonadver- 
sarial hearings in the district office, it was asserted, would be less 
frightening for the applicants, particularly those with meritorious 
cases who were probably the most easily intimidated. Such proceed- 
ings would also allow for swift grants when they were warranted.'** 



^^^ See Memorandum of Solicitor General Suggesting Mootness. Pierre v. 
United Sutes, 454 U.S. 962 (1977) (No. 77-53). 

>»2 See. e.g.. Petition for Certiorari at 18-22, Pierre v. Uniud States. 454 U.S. 962 
(1977) (No. 77-53). 

»" See 43 Fed. Reg. 40,801. 40,802 (1978) (amending 8 C.F.R. pts. 108 and 
236) (affecting excludable aliens and intended to take effect immediately), modified, 43 
Fed. Reg. 48,620 (1978) (in response to litigation, INS stayed the initial rulemaking 
to allow public notice and comment). Also, at the time of the first rulemaking. INS 
promulgated proposed rules intended to work the same changes for deportable 
aliens; that is, to make the immigration court the only venue for asylum claims. See 43 
Fed. Reg. 40,879 (1978). 

'*^ See, e.g., 44 Fed. Reg. 21,253-56 (1979) (summarizing comments received on 
1978 proposed rules). Similar sentiments were expressed with regard to the 1980 
interim rules: 

A number of commenters suggested that all applications for asylum, 
whether filed before or after the institution of exclusion or expulsion 
proceedings, should be decided by the district director. Proceedings 
before the district directors were viewed as less adversarial in nature and 
were providing the applicants with a fi-eer atmosphere within which to 
present their claims. It was pointed out that many applicants have fled 
from countries where the judicial process is suspect and feared by them 
and, they would not feel free to present their claims with the same candor 
that they could in a proceeding before a district director. 
48 Fed. Reg. 5885 (1983). 



284 DAVID A. MARTIN 

Feeling somewhat blindsided, the agency nonetheless largely acqui- 
esced. TTie final rules retained the same two-tier de novo considera- 
tion for deportable aliens, although the rules did insist on 
channeling all claims by excludable aliens into the immigration 
courts.*'^ 

Refugee advocates were not entirely satisfied. Because legisla- 
tion that became the Refugee Act of 1980 was then proceeding 
through Congress, hearings on that bill provided a forum to con- 
tinue the pressure for a more extensive role for the district directors 
in considering asylum claims. Eventually, their lobbying secured a 
measure of congressional support for such changes.'*® INS paid 
heed to these messages, and its interim rules implementing the asy- 
lum provisions of the Refugee Act granted both excludable and 
deportable aliens an opportunity to be heard first in nonadversarial 
proceedings before the district director, provided no charging docu- 
ment had yet issued.''^ If unsuccessful, the claimants retained the 
right to renew their asylum claims before the immigration judges in 
exclusion or deportation proceedings."® With minor changes, these 
interim asylum rules were made final in 1983, retaining the opportu- 
nity for two rounds of de novo consideration. ''® 

B. The Current System 

Current regulations thus establish a complex system — ^rendered 
even more intricate when one takes account of aU the layers of 
review, both mandatory and advisory. The following description, 

»»* S« 44 Fed. Reg. 21,253. 21,258 (1979). These provisions adopted final 
rules requiring that asylum claims by excludable aliens be heard exclusively by 
immigration judges, but permitted deportable aliens not yet in proceedings to file 
before the district director, without prejudice to later consideration of the claim by 
the immigration judge. If an order to show cause had already issued, however, 
asylum and related claims could be heard only by the immigration judge. See id. 

'»6 See, e.g.. The Refugee Act of 1979: Hearings on H.R. 2816 Before the Subcomm. on 
Int'l Operations of the House Comm. on Foreign Affairs, 96th Cong.. 1st Sess.. 89-93 (1979) 
(testimony of Dale F. Swartz, Director of the Alien Rights Law Project of the Lawyers' 
Comm. for Civil Rights Under Law); S. Rep. No. 256. 96th Cong., 1st Sess. 9 (1979) 
(suting Senate Committee's preference for allowing asylum applicants the 
opportunity to have their claims considered outside exclusion or deportation 
proceedings, provided no order to show cause had issued). 

"' See 45 Fed. Reg. 37,392 (1980) (interim regulations adding, inter alia, a new 
8 C.F.R. pt. 208 governing asylum). 

>M See 45 Fed. Reg. 37.392. 37. 395 (1980) (codified 8 C.F.R. § 208.9 (1989)). 

>»9 See 48 Fed. Reg. 5885 (1983) (codified 8 C.F.R § 208.1(b) (1989)) (final nile 
amending the earlier regulations to make clear that jurisdiction vests in the 
immigration judge once a charging document is served, even if the asylum claim vns 
already pending in the district office). 



REFORMING ASYLUM ADJUDICATION 285 

which draws heavily upon the interviews and field observations con- 
ducted for this Article,'"*^ sketches the stages of consideration 
through which an asylum application can proceed. 

1. District Office 

Spontaneous asylum claims, often called "walk-ins" by INS 
officers, receive initial consideration in the district office. To start 
the process, the asylum seeker files the basic application form. Form 
1-589, along with any supporting documents. The four-page form 
asks numerous questions, including queries about past political 
activities, membership in organizations, current whereabouts and 
status of family members, and the applicant's reasons for fearing per- 
secution. Officials sometimes complain that many completed forms 
provide only the scantiest information or seem to follow formulaic 
patterns.'^' Other applications, usually prepared with the assistance 
of counsel, are accompanied by stacks of documents, including both 
affidavits and more generalized information such as news accounts 
and reports fi-om human rights organizations. INS charges no fee 
for filing the 1-589. 

All district offices provide for an interview by an examiner, but 
they vary in the precise arrangements. INS places some emphasis on 
having the interviews conducted by experienced examiners (usually 
at the GS-1 1 grade or higher), who have received special training for 
this task.***^ Offices with heavy walk-in traffic have several such 

•*o Interviews were conducted with: INS Central Office officials (Sept. 1985, 
July 1988, and Aug. 1988); with personnel of the Executive Office of Immigration 
Review (EOIR — the unit of the Justice Department that includes the BIA, 
immigration judges, and certain administrative law judges) in Falls Church, VA (Aug. 
1988); with other Justice Department officials and State Department officials (July 
1988 2md Aug. 1988). I also interviewed immigration judges, INS district office 
officials, private attorneys and voluntary agency representatives, at the following 
locations: Miami (Nov. 1985); San Diego (Feb. 1987); Washington (Oct. 1988); New 
York (Nov. 1988). These visits were supplemented by telephone conversations at 
various times to clarify matters, by conversations at numerous immigration-law 
conferences, and by telephone conversations with officials and attorneys in other 
locations. 

'*' See Dietrich, United States Asylum Policy, in The New Asylum Seekers: 
Refugee Law in the 1980s 67, 70 (D. Martin ed. 1988); P. Fagen, Applying for 
Political Asylum in New York: Law, Policy and Administrative Practice 31 (Occasional 
Paper No. 41, New York Research Program in Inter-American Affairs 1984). 

'^2 The training program has recently been refined under the guidance of a new 
officer in the INS Central Office who formerly worked with the UNHCR. The 
program now includes sessions on the legal provisions and country conditions, and 
usually permits trainees to conduct simulated interviews, followed by critique. Tliese 
measures represent a considerable improvement over earlier practices, when asylum 



286 DAVID A. MARTIN 

examiners, who do nothing but asylum and related refugee work 
during their rotation into this assignment (lasting twelve months or 
more). Smaller offices, however, may of necessity assign these func- 
tions to an examiner who has not had the special training and who 
may devote as little as twenty-five percent of her time to asylum. 

Some districts hold the interview at the time of the filing, but 
most offices with a high volume of asylum traffic schedule interviews 
some weeks or months after receiving the application. New York is 
typical of the latter. When visited for this study in November 1988, 
the New York office had one supervisor and four experienced exam- 
iners who had been doing this work for several years. A fifth exam- 
iner in the Refugee, Asylum and Parole section had less experience 
and so was usually assigned to more routine functions, such as 
processing renewals of asylum status or adjustments to permanent 
residence of persons admitted through the overseas refugee pro- 
gram. The office ordinarily had been able to schedule asylum inter- 
views within sixty days of filing (a date that has significance for work 
authorization purposes),**' but recent increases in applications had 
jeopardized that timing. In an effort to keep up with applications, 
the office began scheduling twenty interviews per day per available 
examiner, although not all of the scheduled applicants were 
expected to show up. The press of business eliminated any opportu- 
nity for examiners to specialize by region of origin of the applicants; 
when one interview was finished, the examiner simply proceeded to a 
central table and picked out the file of the person who had been wait- 
ing the longest. This caseload permitted only about twenty minutes 
per interview, although examiners had discretion to take more time if 
the case required it. 

The interviews usually concentrate on filling in any gaps in the 



training was minimal and examiners had to rely on "on-the-job learning." See 
Immigration 8c Naturauzation Service, Asylum Adjudications: An Evolving 
Concept and RESPONSiBiLrrv for the Immigration and Naturauzation Service SS 
(June 1982) [hereinafter Asylum Adjudications] (a quasi-official internal INS study). 
This 1982 internal study recounted one almost touching vignette that indicates the 
inadequacies that have beset INS adjudications: "One officer said that when she was 
assigned to asylum work, she bought a subscription to Newsweek magazine to 'leam 
more about' countries overseas." Id. at S3 n.*. The officer's initiative is to be 
applauded. But the fact that asylum adjudication might be assigned to someone who 
does not already keep up %vith international affairs at a level represented by weekly 
general circulation magazines is disheartening. That she had to pay for her own 
Newsweek is perhaps even more revealing. 

>^3 See 8 C.F.R. § 274a. lS(d) (1989) (requiring adjudication of an application 
for employment authorization within 60 days or else interim employment 
authorization, good for 120 days, is to be granted). 



REFORMING ASYLUM ADJUDICATION 287 

information presented on the 1-589, primarily with a view toward 
transmitting it all to the State Department for its advisory opinion. 
Typically, the examiner records right on the form in red pen any 
supplementary information developed, although some examiners 
also write out a few sentences of interview notes on a separate sheet 
as well. There is considerable variety in the conduct of the interview, 
depending on the style of the examiner, availability of interpreters 
(or examiners with foreign language skills), and related factors. INS, 
of course, assumes responsibility for making translation available, 
but the interpreter for a particular language may be tied up in immi- 
gration court when needed for an asylum interview in the district 
office. For that reason, INS occasionally relies on family members or 
friends of the applicant for these purposes. Interviews also vary con- 
siderably in thoroughness. One examination I attended in Miami in 
late 1985 (when caseloads were less demanding) lasted nearly an 
hour. The examiner spoke fluent Spanish and was trying with some 
creativity to flesh out the full dimensions of the story told by the 
Nicaraguan appHcant. But another examiner in the same ofhce 
engaged in only perfunctory questioning, concentrating on the 
applicant's manner of entry and other administrative details, rather 
than on the persecution claim. She completed her sessions in about 
fifteen minutes. 

In Miami, at that time, few applicants appeared with counsel. In 
New York in 1988, however, the asylum supervisor estimated that 
perhaps eighty percent of applicants were represented, although asy- 
lum attorneys in New York thought that number a bit high. Attor- 
neys usually play only a bystander's role, partly because examiners 
wish to hear direcdy from the applicant, and partly because, as one 
attorney explained it, not much happens: The interview is "an untax- 
ing experience." Although relations with attorneys appear to be sat- 
isfactory much of the time, most attorneys interviewed recalled 
particular instances where they were thwarted from playing a needed 
role, or occasionally where the examiner was abusive or hostile in 
dealings with the applicant.'"*^ 

The regulations mandate issuance of work authorization to 
"nonfrivolous" asylum applicants, and require that such authoriza- 
tion be provided during all stages of administrative and judicial 



'^'* The American Immigration Lawyers Association (AILA) collected affidavits 
recounting some of these complaints and filed them as part of their opposition to the 
proposed regulations promulgated in August 1987, described tn/ro notes 194-99 and 
accompanying text (affidavits on file with the University of Pennsylvania Law Review). 



288 DAVID A. MARTIN 

review.''** Many attorneys expressed frustration, however, that INS 
rarely issues such authorization at the time of filing, no matter how 
solid the case.'"*^ Instead, the applicant almost always must wait 
until the interview, which may be sixty days away.**^ A substantial 
portion of the interview time is therefore consumed with work 
authorization, including the rather cumbersome physical process 
required to replace the old 1-94 form in the applicant's passport with 
a new one stamped "employment authorized." Obviously, this pre- 
occupation further impairs the opportunity to use the interview to 
examine in detail the particulars of the applicant's persecution 
claim— one important reason why the interviews are often 
"untaxing." 

Not infi-equently, applicants receive no ruling on their asylum 
applications by the time the original employment authorization 
period expires. They may then have considerable difficulty gaining 
extensions, for few district offices have clear channels for making 
such decisions. Under the prodding of the immigration bar, INS is 
now trying to improve arrangements for such extensions, as well as 
for work authorizations for denied applicants who wish to renew 
their applications in immigration court proceedings.'**® The fhiits of 
the interview — the annotated 1-589 with attachments plus any sepa- 

»*5 See 8 C.F.R. fi§ 274a. 12(c)(8), 274a.lS(a) (1989). 

>^ See Memorandum from Richard £. Norton, INS Associate Commissioner for 
Examinations (July 9. 1987), reprinted in 64 Interpreter Releases app. Ill, at 886, 
886-87 (1987) (setting forth standards for deciding on "frivoUty"). In December 
1988. in an apparent attempt to curb the attraction of asylum filings, INS issued 
instructions to field offices stating that the 1-589 was not to be taken per se as a work 
authorization request; a separate application would have to be filed. See INS Clarifies 
Work Authorization Procedures for Asylum Applicants, 66 IroTRPRETER Releases 130, 131 
(1989). INS has also clarified the steps necessary to continue work authorizations if 
the alien wishes to renew the asylum claim in immigration court proceedings. See INS 
Advises on Work Authorization for Denied Asylum Applicants, 65 Interpreter Releases 718 
(1988). 

'^^ Some offices fell much further behind in scheduling interviews and issuing 
work authorizations, prompting Utigation seeking, among other things, to mandate 
compliance with the 60-day limit in the regulations. See Mendez v. Thomburgh, No. 
88-04995 JJH (CD. Cal. amended complaint filed Jan. 30, \9S9), summarized tn Asylum 
Utigation UpdaU, 66 Interpreter Releases 151, 152 (1989). 

'*8 5^^ INS Advises on Work Authorization for Denied Asylum Applicants, 65 
Interpreter Releases 718 (1988); see also Alfaro-Orellana v. lichen, 720 F. Supp. 
792, 798 (N.D. Cal. 1989) (finding that work auihorizations granted to nonfrivolous 
asylum applicants do not terminate upon district office denial of asylum, but continue 
during the entire time that applications are being pursued, up until either 
abandonment or final adverse decision); Doe v. Meese, 690 F. Supp. 1572, 1577 (S.D. 
Tex. 1988) (requiring INS to grant to alien interim employment authorization where 
agency had failed to complete action on her request for work authorization within 
period specified by its own rule); Diaz v. INS. 648 F. Supp. 638, 656 (E.D. Cal. 1986) 



REFORMING ASYLUM ADJUDICATION 289 

rate interview notes — ^are collected for transmission to the State 
Department's Bureau of Human Rights and Humanitarian Affairs 
(BHRHA), under a standard cover sheet containing blanks that allow 
the examiner to provide some additional insights. For example, the 
examiner is asked to characterize the verbal testimony (convincing, 
unconvincing, specific, generalized, etc.) and to provide a prelimi- 
nary assessment (grant, deny, non-committal). State Department 
officers said that examiners often fail to fill out these portions of the 
form. 

After the State Department's views are returned to INS, the 
applicant receives some form of notice, depending on what the dis- 
trict office intends to do with the case. If asylum is to be granted, the 
applicant receives a letter calling her in to complete the paperwork 
for asylee status. If denial is indicated, the applicant receives notice 
of intent to deny, along with a copy of the State Department letter. 
The applicant then has 15 days to rebut or supply additional infor- 
mation.**® "Eager young lawyers," one examiner told me, some- 
times treat this notice as an invitation to provide several pounds of 
additional material. On rare occasions the new information is 
returned to the State Department for further review, but usually the 
matter is simply scheduled for final consideration by an examiner 
after the rebuttal period has passed. In New York, this whole pro- 
cess can last seven or eight months from the time of the interview; 
the process almost surely will require at least four months. For this 
reason, no effort is made to return the file to the original interviewer. 
In any event, she probably would have no independent recollection 
of the case. Several examiners told me that the State Department 
views "count for a lot,"**® although all were aware that they were not 
bound by the Department's position. 

(granting preliminary injunction to aliens prohibiting revocation of work 
authorization until adjudication process for asylum claim had been completed). 

>49 Under a recently implemented procedure described below, if State responds 
with a preprinted sticker indicating simply that it has nothing to add, the district 
office may proceed to a negative decision without issuing a notice of intent to deny. 
See Don't Deny Asylum Cases Just Because of BHRHA "Sticker" Responses, IMS Says, 66 
Interpreter Releases 351 (1989). 

'^ An internal study of INS asylum procedures confirms the great weight 
carried by Sute Department letters. See Asylum Adjudications, supra note 142. at 
57-64. A GAO study based on 1984 advisory opinions found that the Justice 
Department's final decision agreed with the State advice in 96 percent of the cases 
worldwide. See General Accounting Office, Asylum: Uniform Appucation of 
Standards Uncertain — Few Denied Appucants Deported 22, 42 (1987). 



290 DAVID A. MARTIN 

Table I 
Asylum Cases Filed in INS District Offices 

Fiscal year Total received 

1984 24,291 

1985 16.622 

1986 18.889 

1987 26.107 

1988 60,736 

1989 101.679 
Source: Immigration and Naturalization Service. 

Table I provides statistics on the rising caseloads of the district 
offices, and Table II shows approval rates by nationality for fiscal 
year 1988 and cumulatively for the last five years. The statistics show 
the number of "cases." Because a case may represent applications 
for several members of a nuclear family, who are treated together in 
accordance with IN A § 208(c), actual numbers of asylum seekers are 
higher than what is shown in the Tables. Moreover, the tables do 
not include applicants who apply only before the immigration 
judges. 



REFORMING ASYLUM ADJUDICATION 



291 



Table II 

Grants and Denials by Nationality 

Asylum Appucations Filed in INS District Offices 

June 198S-Septeinber 1988 





Cumulative* 






Fiscal year 1988 






Approval 






Approval 










Rate for 






Rate for 










Cases 


Cases 


Cases 


Cases 


Cases 


Cased 


r^ses 


Country 


Decided 


Granted 


Denied 


Decided 


Granted 


Denied 


Pending 


total" 


27.8% 


28.416 


73.753 


39.1% 


5.531 


8.582 


73.109 


Iran 


61.7% 


12.459 


7.727 


75.0% 


764 


254 


1,373 


Romania 


61.4% 


895 


562 


82.9% 


345 


71 


341 


Czechoslovakia 


44.7% 


123 


152 


44.8% 


13 


16 


72 


Syria 


S9.4% 


186 


285 


65.7% 


25 


13 


100 


Ethiopia 


39.0% 


1.340 


2.089 


77.0% 


441 


131 


525 


Poland 


38.0% 


2.686 


4,365 


53.7% 


433 


373 


2,099 


Afghanistan 


37.0% 


402 


684 


39.5% 


36 


55 


91 


China 


32.6% 


167 


345 


69.7% 


60 


26 


142 


Vietnam 


31.3% 


68 


149 


80.0% 


8 


2 


9 


Hungary 


30.0% 


175 


407 


28.9% 


24 


59 


172 


Nicaragua 


27.9% 


7.255 


18.688 


53.1% 


2,786 


2,455 


21,054 


Uganda 


26.0% 


91 


258 


51.7% 


15 


14 


34 


Somalia 


24.1% 


143 


449 


67.9% 


55 


26 


122 


Philippines 


18.0% 


82 


372 


10.0% 


4 


36 


94 


Cuba 


13.3% 


321 


2.080 


31.9% 


SO 


64 


13.873 


Yugoslavia 


13.3% 


53 


344 


9.2% 


6 


59 


176 


Pakistan 


13.0% 


63 


420 


57.8% 


33 


24 


89 


Liberia 


10.5% 


30 


255 


15.0% 


3 


17 


32 


Lebanon 


7.0% 


113 


1.499 


36.6% 


56 


97 


222 


Honduras 


4.3% 


18 


398 


7.4% 


10 


125 


512 


El Salvador 


2.7% 


667 


23.805 


2.7% 


110 


3.822 


24,375 


Guatemala 


2.1% 


45 


2.086 


5.0% 


24 


447 


6,191 


Haiti 


2.0% 


36 


1.713 


31.5% 


6 


13 


399 


Egypt 


1.0% 


8 


754 


0.5% 


1 


19 


32 


India 


1.0% 


4 


379 


15.0% 


3 


17 


29 


Bangladesh 


0.2% 


1 


416 


16.6% 


1 


5 


10 



'Since May 1983. INS has kept asylum statistics by number of cases; each case, or 

application, may include more than one individual. The uble cumulates the data 

only from the time this statistical uniformity was established. 

" The toul includes all nationalities, not only those designated here. 

Source: Immigration and Naturalization Service, as compiled in Refugee Reports, 

Dec. 16, 1988. at 14. 



2. Immigration Court 
Claimants whose applications were denied in the district office 



292 DAVID A. MARTIN 

may renew their asylum applications before the immigration judges 
when deportation or exclusion proceedings begin. If the alien 
applies for asylum only after those proceedings have been initiated, 
immigration court will provide the only available forum. '^* In either 
case, the process is virtually identical. It is initiated by filing Form I- 
589, along with any accompanying documents, with the docket 
clerk. *^^ The clerk then forwards a copy to the State Department. 
Although the regulations seem to preclude (with limited exceptions) 
State Department referral if the district oflfice earlier received an 
advisory letter,'*' docket clerks now routinely transfer the files with- 
out checking for earlier letters. Not only is this arrangement less 
cumbersome for the clerk, but it also fits better with the desire of the 
Elxecutive Office of Immigration Review (EOIR) to assure the inde- 
pendence of the present proceedings from earlier INS consideration. 
State Department officers reported that they received less complete 
files from the immigration courts, perhaps because some docket 
clerks resist mailing lengthy documentary attachments. In any event. 
State clearly receives nothing equivalent to the district office exam- 
iner's notes. Some judges even provide a cover sheet pointedly stat- 
ing that no assessment of credibility or other review has been 
performed before transmission to the State Department. 



>5« See 8 C.F.R. §§ 208.3(b), 208.9 (1989). Many aliens first express a wish to 
apply for asylum when the case comes before the immigration judge on "master 
calendar," a short while after the order to show cause (which initiates deportation 
proceedings) has been served on the alien. At the master calendar hearing, the alien 
pleads to the order to show cause and makes known any defenses or applications for 
reUef from deportation. The overwhelming majority of asylimi applicants admit 
deportability; asylum is then usually the only issue in the deportation proceeding. 

152 Technically, an applicant in the immigration court is seeking the benefits of 
both asylum under INA § 208 and withholding under § 243(h), whereas examiners in 
the district office can award only asylum under § 208. This distinction makes 
virtually no practical difference, and Form 1-589 is identical in both settings. 

It is also possible that an asylum claim will be lodged later in the process. For 
example, the regulations still provide that an alien be expressly advised of her right 
to apply for withholding at the time the immigration judge designates a country of 
deportation. See 8 C.F.R. § 242.17(c) (1989). The alien can then receive 10 days to 
fill out the 1-589 and thereby initiate consideration of a persecution claim. See id. An 
alien can also apply for asylum after the issuance of a deportation order by filing a 
motion to reopen, but the alien then carries the additional burden of explaining the 
failure to apply earlier. See 8 C.F.R. § 208.1 1 (1989). This burden can be substantial. 
See, eg.. INS v. Abudu. 485 U.S. 94, 109-1 10 (1988) (analogizing a motion to reopen 
to a motion for a new criminal trial on the basis of newly discovered evidence — a 
motion in which the moving party "bears a heavy burden"). If reopening is granted, 
the matter will return to the immigration judge for consideration. 5^^ 8 C.F.R. 
§§ 103.5, 242.22 (1989). 

'" See 8 C.F.R. §§ 208.7, 208.10(b) (1989). 



REFORMING ASYLUM ADJUDICATION 293 

After State's views are received, the case can be calendared for a 
hearing. The timing varies depending on caseload, but delays of a 
year are not unknown. Asylum cases receive no priority in calendar- 
ing unless the applicant is detained, in which case the judges place a 
high priority on prompt adjudication. Detention is more likely in 
exclusion cases than in deportation.'^'* The immigration court in 
New York has worked out arrangements with attorneys to permit 
time for interviews with clients when the clients are brought in to the 
court facility in Brooklyn, thus obviating frequent trips to the remote 
detention facility.*^* 

Procedure in court conforms, by and large, to a standard adver- 
sarial model of a trial-type proceeding. Most asylum seekers, at least 
in the districts with high volume, are now represented in immig^- 
tion court by counsel or accredited nonattomey representatives.**® 
The burden of proof is on the alien. Counsel will usually elicit the 
key particulars of the story from the client on direct examination and 
will also offer available supporting materials, often derived from the 
reports of human-rights NGOs such as Amnesty International, 
Americas Watch, or the International Commission of Jurists.**' The 



»M 5» 8 C.F.R. §§ 212.5. 235.3 (1988). The background of the detention policy 
is set forth in T. Aleinikoff & D. Martin, n^a note 3, at 722-24. See also Amanullah 
V. Nelson, 811 F.2d 1, 4-8 (1st Cir. 1987); Singh v. Nelson, 623 F. Supp. 545, 556 
(S.D.N.Y. 1985) (generally upholding detention of asylum seekers, in part as a 
deterrent measure). For critiques of detention of asylum seekers, see Helton, si^a 
note 1 16; Levy, Detention in the Asylum Context, 44 U. Pnr. L. Rev. 297 (1983). 

•55 Detention arrangements bring frequent complaints from applicants' 
attorneys, particularly when detention is carried out by private contractors, as has 
sometimes been the case, for example, in New York City. I was told that "these are 
guys who usually guard construction sites," that some of them "know nothing" about 
American Correaional Association accreditation standards, and that they often make 
life quite difficult for attorneys who are simply seeking access to their clients. 
Interview with Arthur Helton. John Assadi, and Jeff Heller, Attorneys with the 
Lawyers' Comm. for Human Rights, in New York City (Nov. 15, 1988); see also 
General Accountinc Office, Criminal Auens — INS' Detention and Deportation 
AcnvmEs in the New York Cmr Area 24-27 (1986). 

»56 See 8 C.F.R. § 292.2 (1988) (providing for accreditation of such 
representatives). Part 292(e) requires district directors to maintain a list of fi-ee legal 
services programs available in the area. These lists can be somewhat misleading, 
both because the lists get out of date and because some of the organizations listed 
have in recent years become more selective in accepting cases. 

' 57 The closest analogues in this country to the documentation centers used by 
European adjudicators are the private documenution centers put together by 
refugee advocacy organizations. Many specialize by region. See, e.g., New 
Documentation Center Announced for Salvadoran Asylum Cases, 60 Interpreter Releases 
975 (1983) (reporting on new center, established by the American Civil Liberties 
Union Fund of the National Capitol Area and the Center for National Security 
Studies, gathering information on EU Salvador). At least one commercial 



294 DAVID A. MARTIN 

INS trial attorney then cross-examines. In busy districts, trial attor- 
neys have little time to prepare the cases. Sometimes they are only 
able to review the file for the first time while direct examination is 
proceeding. Moreover, trial attorneys are not generally expected to 
develop extensive additional information or other sources of 
evidence. 

This insufficiency in preparation time and resources results in 
several disadvantages, which are compounded by the failure to 
assign clear responsibility for the development of other sources of 
information. Cross-examination is impoverished under these cir- 
cumstances. One trial attorney stated ruefully during her interview 
that she necessarily does what she was always taught in law school 
not to do: She asks questions when she has no idea where the 
answers might lead. Sometimes a useful line of inquiry develops. 
Often it does not. All she can do is probe apparent soft spots and 
inconsistencies in the story. Given so litde to work with, her incen- 
tives are to magnify the weaknesses in the testimony even if there 
might be an innocent explanation. 

Second, an early review of the file might have revealed the need 
for additional and specific information on country conditions. 
Review on the day of the hearing will be too late, even if the informa- 
tion would have been relatively accessible. An example may be help- 
fill in illustrating this point. Suppose the applicant claims that she 
fled to avoid forced conscription into the army or a guerrilla unit. 
Cross-examination can do litde to explore whether such dragooning 
really takes place; it can only probe the wellsprings of this pardcular 
alien's belief. If the State Department response said nothing about 
the issue (perhaps because the initial papers did not make sufficienUy 
dear that this was the basis of the claim), the record may contain no 
useful general information on this crucial question. The applicant's 
assertions will therefore stand "uncontroverted," whatever may be 
the real state of affairs in the home country. BIA member Michael 
Heilman expressed particular concern about this inadequacy, espe- 
cially given the Board's strict adherence to the requirement that 
decisions be based on evidence of record in the particular 
proceeding.'*® 



organization has become involved. The Data Center in Oakland, California recendy 
mailed a brochure to members of the American Immigration Lawyers Association 
advertising its Political Asylum Documentation Service, available for $50 per hour of 
search time, plus photocopying and postage. 

158 Telephone interview with Michael Heilman, Member, Board of Immigration 
Appeals (Sept. 22, 1988). 



REFORMING ASYLUM ADJUDICATION 295 

Some trial attorneys also reflected thoughtfully on the wider 
implications of applying the adversarial system to these matters. The 
government's real interest will not always be to oppose the claimant; 
some of the applicants deserve asylum. But the INS attorney may 
have litde idea which type of case is before her until well into the 
proceedings. Moreover, as one attorney told me, even when it 
appears to be a strong case, her instincts (and perhaps her inevitable 
role under this structure) lead him to react in a particular way: 
"When it's there in the courtroom, I'm 'agin' it."'*® 

There are currendy seventy-five immigration judges. As Table 
III indicates, asylum cases have risen from ten percent of immigra- 
tion court caseload in 1985 to over thirteen percent in 1988. 
Because asylum cases tend to present the most difficult and challeng- 
ing issues appearing in a judge's caseload, however, they occupy a 
much higher percentage of actual work time. Statistics on disposi- 
tions have only recendy been maintained. In fiscal year 1988, immi- 
gration judges received 11,025 new asylum cases and disposed of 
them as follows: 1,647 cases were granted (representing 2,276 indi- 
viduals); while 5,626 were denied. The overall grant rate was twenty- 
three percent. A total of 4,364 cases were pending in immigration 
courts at year's end.*®® 

•*® Interview with former trial attorney, in New York City (Nov. 14, 1988). 

>60 Executive Ofhce for Immigration Review (EOIR), Asylum Statistics 
[Immigration Judges] for Period 10/1/87-10/1/88. The slight discrepancy 
between these numbers for asylum cases received and those in Table III is 
unexplained. Until very recently, EOIR published only limited statistics on asylum 
cases and did not reveal grant and denial rates by nationality. In the spring of 1989, 
however, it decided to compile such figures and make them available to the public. 
They show, for example, a noticeably higher grant rate for Salvadorans (12 percent) 
and a lower one for Nicaraguans (37.4 percent) than the comparable rates in the INS 
district offices, which arc shown sttpra Table II following note 150. The complete set 
of EOIR statistics, with details by nationality, appears in Refugee Reports, May 19, 
1989. at 5. 



296 DAVID A. MARTIN 

Table III 
Cases Received, Immigration Judges 



Fiscal 






Motion to 




Asylum 


Percent 


Year 


Deportation 


Exclusion 


Reopen 


Total' 


Cases'* 


Asylum 


1985 


102,044 


9,122 


3,521 


1 14,687 


11,663 


10.2% 


1986 


89.680 


9,576 


3,555 


102,811 


11,156 


10.9% 


1987 


64,133 


9,178 


2,711 


76,022 


8,659 


11.4% 


1988 


71,308 


10,167 


2.387 


83,862 


11,200 


13.4% 


1989 


118,906 


14,020 


2.398 


135,324 


20,331 


15.0% 



* Immigration judges also hear a fourth category of cases, involving release on bond. 

But because asylum is not an issue in such proceedings, these totals omit that 

category in order to obtain a meaningful base for comparison. 

** EOIR does not break down asylum receipts according to the type of procedure 

(deportation, exclusion, motion to reopen) in which the application is received. The 

percentage is therefore stated as a portion of total combined caseload in those three 

categories for the year. 

Source: Telephone interviews with Gerald Hunvitz. Coimsel to the Director. EOIR 

(Mar. 8. 1989. Mar. 28. 1990). 



3. State Department Role 

The State Department is required by statute to publish annual 
reports on human rights conditions in all foreign countries.*®* This 
requirement derived from congressional efforts to strengthen human 
rights policy during the Kissinger era at the State Department, rather 
than from any concern for asylum proceedings. Nevertheless, the 
reports have proven to be extremely useful in asylum adjudications, 
both in the United States and around the world. Every district office 
asylum unit and every immigration judge receives a copy, and I have 
seen well-thumbed issues in the offices of asylum adjudicators in 
Western Europe and Canada — and indeed in the offices of UNHCR 
and refugee advocacy organizations. 

The controversial portion of the State Department's role relates 
to its preparation of advice letters in individual asylum cases. The 
Bureau of Human Rights and Humanitarian Affairs (BHRHA) per- 
forms the central functions in the asylum advice process. Its asylum 
unit, headed by a career foreign service officer, is largely staffed by 
retired foreign service officers, doing part-time work under contract. 
Such officers are able, to a considerable extent, to specialize by 
region of origin. In the summer of 1988, for example, one officer 
assumed responsibility for claims from Eastern Europe and some 

'61 See 22 U.S.C. i 2151n(d) (1988). 



REFORMING ASYLUM ADJUDICATION 297 

East Asian countries. Another concentrated on the Near Elast and 
South Asia. Three officers were responsible for Central and South 
America. 

When files arrive from either district offices or immigration 
courts, they are logged into a central bureau computer and assigned 
to the appropriate officer for review. Until eariy 1988, BHRHA 
returned an opinion letter in each case, many of them standard form 
letters, announcing whether the Department beheved that the partic- 
ular applicant had a well-founded fear of persecution.*®^ In almost 
all cases, BHRHA officers have initially drafted the letters after sim- 
ply reading through the file. Very few cases have stimulated further 
specific research. After drafting, the letter is cleared with the appro- 
priate country desk in the regional bureaus of the State Department. 
Critics have sometimes targeted the regional bureaus as the source 
of political bias in the letters. State Department officials, however, 
deny that diplomatic commitments to other nations have ever 
entered into die advice-letter process.*®' Moreover, as the number 
of cases has expanded, the regional bureau clearance process has 
become more and more routine. Apparendy, regional bureaus often 
grant blanket clearances, relying on the BHRHA officer to bring any 
unusual cases requiring more thorough scrudny to their attendon. 

In February 1988, State introduced a new system. INS and 
EOIR stiD send all files to State as before, but the Department no 
longer invariably returns an advice letter. Instead, its response takes 
one of three forms. First and most common, a sdcker is affixed to 
the returning papers, stadng that the Department has nothing to add 
and referring generally to the latest human rights country reports.*®* 
Second, the officer might return the file without an individualized 
communicadon, but instead with a general update sheet. Those 
sheets report, for example, a change of government since the last 
country report was written, or provide more detailed information 

>62 Form letters have often drawn criticism, but volume made such an approach 
virtually unavoidable. See generally In re Vigil, Interim Dec. No. 3050, slip op. at 12 
(BIA Mar. 17, 1988) (finding form deporution letter not to constitute an error 
because it is not binding and immigration judge may determine weight it deserves). 

'6' See Burke, Compassion versus Self-interest: Who Should Be Given Asylum in the 
United States?, 8 Futcher Forum 31 1, 320-22 (1984); Dietrich, supra note 141. at 67- 
72. 

'" The text of the sticker is reprinted in 8 AILA Monthly Maiunc 1 18 (1989) 
(reporting further on INS cable clarifying that no "notice of intent to deny" is 
necessary when the Sute Department responds with a sticker). Because examiners 
apparently tended to treat stickers as negative opinions, INS recently had to send 
further guidance to its offices emphasizing that stickers are not to be considered as 
"adverse evidence." 66 Interpreter Releases 351 (1989). 



298 DAVID A. MARTIN 

than the country report about specific issues (such as the treatment 
of a particular religious group or punishments for those who left the 
country without exit permission). TTiird, the Department still some- 
times returns a more detailed advisory letter in cases where it has 
something specific to add.'®* 

The new sticker system has received mixed reviews. Several 
people interviewed, including William Robie, the Chief Immigration 
Judge, find the change a big improvement. '®® The stickers make it 
clearer than the old form letters did that the Department has nothing 
to add. It is then more clearly left to the immigration judge to decide 
based on the record — ^which can include the annual State Depart- 
ment country report and any generic update sheets that are avail- 
able. Others, especially district office examiners, but also some 
immigration judges, regard the stickers as a dereliction of duty; they 
want more guidance from State. One State Department officer (who 
also dislikes the sticker system) reported frequent receipt of files 
from district offices with a maiiung on the cover sheet: "No 
sticker."'®' 

Asylum applicants have frequently challenged the use and accu- 
racy of State Department letters, or have sought access to their 
authors for cross-examination. Although some letters have drawn 
sharp judicial criticism,'®^ by and large the courts have approved the 
practice, even in the immig^tion court setting.'®® In one such case. 
Judge Henry Friendly, in dictum, advised certain changes that would 
minimize any due process problems arising fr-om introduction of the 
letters without making their authors available for cross-examination. 
In the future, she recommended, such letters should be confined to 
dealing with "legislative facts" such as the extent of persecution in 
the home country, and refrain from recommending an outcome in 
the specific case. If so modified. Judge Friendly suggested, they 
would deal with matters "on which the safeguards of confrontation 
and cross-examination are not required and on which the IJ needs all 



>65 See 53 Fed. Reg. 2893 (1988) (notice announcing the new system). 

166 Interview with William Robie. in Falls Church, Va. (Aug. 12. 1988). 

'6' Interview with Christopher Squier. in Washington. D.C. (Aug. 12. 1988). 

>68 See, e.g.. Paul v. INS, 521 F.2d 194. 199-200 (5th Cir. 1975); Kasravi v. INS, 
400 F.2d 675. 676-77 Sc 677 n.l (9th Cir. 1968). 

»69 See, e.g., McLeod v. INS. 802 F.2d 89. 95 n.8 (3d Cir. 1986). Hotel and 
Restaurant Employees Union. Local 25 v. Smidi, 594 F. Supp. 502. 510-14 (D.D.C. 
1984). aff'd by an equaUy divided court, 846 F.2d 1499 (D.C. Cir. 1988). For BIA 
descriptions of the appropriate use of Sute Department advice, see In re Vigil. 
Intenm Dec. No. 3050. slip. op. at 12 (BIA Mar. 17. 1988); In re ExUus. 18 I. & N. 
Dec. 276. 278-80 (1982). 



REFORMING ASYLUM ADJUDICATION 299 

the help she can get."'^^ For roughly 13 years after Judge Friendly's 
admonition, the Department persisted in sending letters that com- 
mented on the specific case, i.e., on adjudicative fiacts. The new 
sticker system (ironically, adopted primarily for budgetary reasons) 
now brings practice more in line with the scheme Judge Friendly 
envisioned. 

4. Administrative Review 
a. The Board of Immigration Appeals 

There is no appeal from the district office decision, although 
renewal in immigration court obviously prpvides an opportunity to 
secure relief if the original denial was unjustified. Decisions in immi- 
gration court, however, are appealable to the Board of Immigration 
Appeals (BIA), under the standard procedures allowing review of 
decisions in exclusion and deportation cases. Both the applicant and 
the INS can appeal, although appeals by the latter are far less 
frequent. 

The Board, created by Justice Department regulations rather 
than by statute, consists of five members and conducts business at its 
headquarters in Falls Church, Virginia. It hears appeals in exclusion 
and deportation cases, and also reviews a variety of other immigra- 
tion-related decisions.'^' Exact statistics are unavailable, but 
informed guesses place asylum at about one-quarter of the BIA's 
caseload. Owing to their greater complexity, however, asylum cases 
occupy about half its work-time. 

To cope with a rising workload, new regulations, enacted in 
1988, authorized temporary assignments of immigration judges to sit 
with the Board, and for the first time allowed consideration of cases 
by three-member panels. *^^ Oral argument is possible if the Board 
approves, but such approval remains rare. It is far more common for 
the Board to consider the matter on the briefs alone. Each appeal is 
decided by opinion, but only a small fraction of those opinions are 
published as precedent decisions. *'* The BIA appeal process can 
consume considerable amounts of time, although precise statistics 

•70 Zamora v. INS, 554 F.2d 1055, 1062 (2d Cir. 1976). 

•" See Legomsky, Forum Choices for the Review ofAgertcy Adjudication: A Study of the 
Immigration Process, 71 Iowa L. Rev. 1297, 1302-10 (1986). 

«72 See 53 Fed. Reg. 15,659 (1988) (codified at 8 C.F.R. S 3.1(a)(1) (1989)). 

• 75 For a useful look at BIA practice, see Note, A Refugee by Any Other Name: An 
Examination of the Board of Immigration Appeals' Actions in Asylum Cases, 75 Va. L. Rev. 68 1 
(1989). 



300 DAVID A. MARTIN 

are not maintained. The greatest delays derive from backlogs in typ- 
ing the transcript of the hearing; periods of eight to twelve months 
are not uncommon. The Board then requires several additional 
weeks or months before issuing a decision.*^'* Cases involving an 
applicant in detention, however, receive priority. In these instances, 
the transcript can be prepared in a matter of weeks, and the Board 
will also expedite its own decision process. 

Table IV 
Cases Received, Board of Immigration Appeals 

Fiscal Year Total Cases 

1985 4,911 

1986 8,608 

1987 8,204 

1988 10,191 

1989 11.186 

Note: The EOIR computer system did not separately code asylum appeals before 
1989, but observers agreed that asylum cases are appealed more often than other 
decisions by immigration judges. The percentage of asylum cases should thus be 
considerably higher than the p>ercentage of receipts for immigration judges, svfrra 
Table III following note 160. 

Source: Telephone interviews with Gerald Hurwitz, Coimsel to the Director, 
EOIR (Mar. 8. 1989. Mar. 28. 1990). 

b. The Asylum Policy and Review Unit 

The Asylum Policy and Review Unit (APRU) was established in 
the Office of Legal Policy of the Department of Justice in 1987, at 
least pardy in response to the Medvid incident (the forcible return of 
a Ukrainian seaman to a Soviet ship docked in Lx>uisiana).'^^ Its gen- 
esis also reflects the dissatisfaction of the Justice Department under 
Attorney General Meese with the tenor of the handling of certain 
cases by the State Department and INS. An APRU official com- 
plained, for example, that State failed to stay sufficiendy "up to date" 

"■* See Legomsky, supra note 171, at 1331 (reporting rough estimate of three 
months' mean for BIA disposition — presumably from the time the BIA actually 
receives the case; that is, after the transcript and briefs are received). Transcript 
delays are so cosdy that the system should explore other means of presenting an 
adequate record on appeal. For an innovative suggestion of using videotapes for 
these purposes in asylum cases, see E. Ratushny, supra note 64, at 57-58. 

"5 The regulation officially establishing the Unit appears at 52 Fed. Reg. 
11,043 (1987) (codified at 28 C.F.R. { 0.23b (1989)). See also New Asylum Policy and 
Review Unit Created, 64 Interpreter Releases 439-40 (1987); Attorney General 
Announces New Asylum Policy Unit, 64 Interpreter Releases 472-73 (1987). 



REFORMING ASYLUM ADJUDICATION 301 

on developments in some countries, '^^ Critics of APRU provide a 
harsher assessment of its origin. They believe it was created to 
assure a higher grant rate for persons fleeing Eastern Europe or 
other Marxist countries — apparently a matter of strong concern to 
the Department under Mr. Meese.*^^ 

APRU receives the whole file in every case denied in the district 
oflices, and a copy of the approval letter in g^ranted cases. Either 
way, APRU almost always receives the material after the alien has 
been notified of the result. The office carefully keeps its distance 
fi"om cases being considered in EOIR (by the immigration judges or 
the BIA), in order to honor the latter's quasi-judicial independence. 
In a handful of cases, apparendy, an applicant rejected in EOIR has 
been given a new round of review in the, district office as a result of 
APRU concern. APRU is not stricdy an appellate body. Asylum 
applicants do not initiate its consideration, although knowledgeable 
lawyers are now becoming more aware of the office's role and of 
course cannot be prevented fi-om writing with concerns about alleg- 
edly unjustified denials. 

Most of APRU's review work is performed by three attorneys in 
the office. If they believe an application was improperly denied, or 
spot trends indicating undue restrictiveness with respect to certain 
groups, APRU makes its concerns known to the Central Office of 
INS. Sometimes this results in reopening and correction in the dis- 
trict office. At the time of interviews (July 1988), however, the Dep- 
uty Director expressed concern about whether messages 
communicated in this way adequately get through to the districts. 
She further described APRU's role as a "safety valve," assuring that 
persons at risk are not wrongfully sent home; the office's individual 
case review serves mainly to spot egregious cases. About forty such 
cases had been pursued with vigor in the district offices.*^® INS, 
however, believes that APRU's functions are duplicative, and INS 

'^6 Inierview with Robert Charles Hill, Deputy Director, APRU, in Washington, 
D.C. (July 28, 1988). 

'7^ See, e.g., Pear, supra note 40 (reporting on the consideration of regulations 
that would provide a presumption favoring asylum for persons from "totaliurian" 
but not from "authoritarian" countries; the proposal later drew considerable 
criticism and was never ofRcially made public). 

"8 Interview with Robert Charles Hill, Deputy Director, APRU, in Washington. 
D.C. (July 28. 1988). As of April 5, 1989. the Director reported: "APRU has worked 
with INS to obtain the reversal on approximately 40 cases. In another approximately 
40 cases, no agreement could be reached with INS. and APRU recommended, and 
the Deputy Attorney General approved, grants of asylum. In addition, approximately 
50 cases are currently being discussed by APRU with INS." Letter from Henry L. 
Curry to David A. Martin (Apr. 5. 1989). 



302 DAVID A. MARTIN 

Commissioner Nelson urged Attorney General Thomburgh to abol- 
ish the office, allowing INS to allocate APRU's annual budget to 
other parts of the adjudication system.'^® 

5. Judicial Review 

Applicants ordinarily secure judicial review of asylum denials in 
accordance with the regular arrangements for review of exclusion or 
deportation orders under IN A § 106.'®° Elxclusion cases therefore 
proceed to district court on a habeas corpus petition; deportation 
cases are heard in the court of appeals based on a petition for review. 
In practice, this distinction makes litde or no difference in the opera- 
tive scope of review. Courts review denials of the mandatory protec- 
tions of § 243(h) to check that the ruling was based on "substantial 
evidence."*®' Denials of asylum under § 208 are subject to a bifur- 
cated standard of review. The "substantial evidence" test applies to 
factual determinations that underlie thegudgment as to whether the 
person has a well-founded fear of persecution, but if asylum is 
denied in the exercise of discretion, that denisd is reviewed only for 
"abuse of discretion" — intended as a more deferential standard.'®^ 
Whatever the precise formula, the actual vigor of scrutiny covers a 
wide range, firom highly deferential to highly demanding.*®' 

Deportation is automatically stayed once a petition for review ii 
served on INS.*** Stays are not automatic in exclusion cases or while 

> 79 5m 66 iNTCRPRrrER Releases S (1989) (asserting that from Apr. 1987 
through Dec. 1988, APRU cost the INS appropriation $750,000). 

•80 8 U.S.C. § 1 105a (1988). Some cases appear to hold open the possibility of 
judicial review in district court under the Administrative Procediu-e Act (APA) of 
denials in the district office, but the better view {>ost|x>nes court consideration until 
immigration court and BIA remedies have been exhausted. See, e.g., Chen Chaun-Fa 
V. Kiley. 459 F. Supp. 762. 765 (S.D.N.Y. 1978). 

'81 See, e.g.. Chavarria v. Department of Justice. 722 F.2d 666, 670 (11th Cir. 
1984); McMuUen v. INS. 658 F.2d 1312. 1316-17 (9th Cir. 1981). Bu/ «f Marroquin- 
Manriquez v. INS. 699 F.2d 129. 133 (3d Cir. 1983) (suting that an "abuse of 
discretion" standard should be used, because of the "necessary application of 
expertise in the determination that a fear of persecution is well-founded"), cert, denied, 
467 U.S. 1259 (1984). 

>82 See. e.g. Cruz-Lopez v. INS. 802 F.2d 1518. 1519 n.l (4th Cir. 1986); Vides- 
Vides v. INS. 783 F.2d 1463. 1466 (9th Cir. 1986); Carvajal-Munoz v. INS. 743 F.2d 
562, 567-68 (7th Cir. 1984). 

'8» Compare Turcios v. INS. 821 F.2d 1396, 1399 (9th Cir. 1987) and Damaize- 
Job V. INS. 787 F.2d 1332. 1338 (9th Cir. 1986) (close review of immigration judge's 
credibility rulings) with Diaz-Escobar v. INS. 782 F.2d 1488. 1492 (9th Cir. 1986) and 
Sarvia-Quintanilla v. INS. 767 F.2d 1387. 1395 (9th Cir. 1985) (great deference to 
agency's credibility determinations). 

>U See INA § 106(a)(3). 8 U.S.C. $ 1 105a(a)(3) (1988). 



REFORMING ASYLUM ADJUDICATION 303 

a motion to reopen is pending, but in light of the possible effects of 
an erroneous removal, district courts have been quite hospitable to 
the issuance of a stay.'®* Given federal court caseloads, pursuing 
judicial review can considerably lengthen an applicant's stay in the 
United States. But as Table V indicates, only a small proportion of 
asylum applicants pursue direct review in court. The information in 
the Table is somewhat misleading, however, as class actions or other 
suits seeking broadly applicable injunctive relief, rather than direct 
review in single cases, have provided the setting for some of the most 
important judicial rulings on asylum processing. '®® Some of these 
cases have resulted in multi-year stays of removals and in orders 
necessitating reopening of hundreds of proceedings. The Director 
of the Justice Department's Office of Immigration Litigation reports 
that asylum issues "take a huge portion of our time."'®^ 

»85 See, e.g., Bazrafshan v. Pomcroy, 587 F. Supp. 498. 501 (D.N.J. 1984) (suting 
that we "cannot let the rigors of administrative procedure reshape our ideas about 
life and death"). 

'** See cases cited supra note 3. 

'*'' Telephone interview with Robert Bombaugh, Director of the Justice 
Department's Office of Immigration Litigation (Mar. 8, 1989). 



304 



DAVID A. MARTIN 



Table V 
Direct Review in Federal Court 





Total 












applications 


No. 


presenting 


Percent 


Fiscal year 


for review 


asylum issue 


asylum 






Deportation 


a 






1984 


425 






116 


27.3% 


1985 


427 






116 


27.2% 


1986 


396 






115 


29.0% 


1987 


161 






43 


26.7% 


1988 


179 






49 


27.4% 


1989 


275 


Exclusion** 




120 


43.6% 


1984 


18 






3 


16.7% 


1985 


25 






19 


76.0% 


1986 


19 






7 


36.8% 


1987 


18 






6 


33.3% 


1988 


12 






9 


75.0% 


1989 


66 






57 


86.4% 



* These statistics probably undercount asylum cases. Petitions for review are logged 
in on the Office of Immigration Litigation OIL statistical system at the time of filing, 
but it may not be apparent until bter stages that an asylum issue is presented. An 
effort b made to go back and correct or amplify the entries, but a few cases are 
overlooked in the process. That correction process is still underway for the most 
recent fiscal years; it is therefore likely that those years' asylum sutistics will rise, 
perhaps substantially. 

** Exclusion cases are almost certainly undercounted. Review is obtained by petition 
for habeas corpus in the district courts, and the local U.S. Attorney's office represents 
the government. Not all such offices report full statistics to OIL. 
Source: Telephone interviews with Robert Bombaugh, Director of the Office of 
Immigration Litigation, Dept. of Justice (Mar. 8, 1989, May 10, 1990). 



6. Actual Deportations 

It is difRcult to obtain precise information on actual removals of 
unsuccessful asylum seekers, but by all accounts, the numbers are 
quite low. A 1984 GAO study found final deportation orders issued 
in only 3.5 percent of cases of persons initially denied asylum, and 
over half of those individuals apparently had not yet been removed. 
Another one percent had left on their own. Although a very high 
percentage of the sample of 21,032 aliens were in an uncertain status 
(and some might have left), these numbers are disturbing, particu- 
larly because actual deportations are essential to send any kind of 



REFORMING ASYLUM ADJUDICATION 305 

deterrent message to persons in the home countries contemplating a 
journey to the United States to apply for asylum.*®® 

Deportations falter at two stages. First, applicants denied asy- 
lum in the district office will receive orders to show cause (thus initi- 
ating deportation proceedings) only if the investigations section 
completes the paperwork. These sections are overburdened, and 
failed asylum applicants do not occupy a high enforcement priority. 
Second, when a deportation order becomes final, the respondents 
are usually promptly served with a notice to leave the country. But if 
they fail to do so, enforced deportation will occur only if INS takes 
the initiative to locate and apprehend the person. Again, failed asy- 
lum seekers occupy a low enforcement priority.*®^ 

There are exceptions to these patterns. If the individual is held 
in detention, a final deportation or exclusion order will almost 
always result in prompt removal. Moreover, several officers in dis- 
trict offices reported to me that voluntary surrender for deportation 
picks up just before holidays. Apparently, the individuals are ready 
to go home, and they present themselves to INS because the agency 
is likely then to pay the costs of transportation. It is not surprising 
that such behavior by unsuccessful asylum applicants evokes cyni- 
cism on the part of the officers.'^ 

7. The Role of the UNHCR 

In the late 1970s, the State Department made arrangements 
with UNHCR for its review of all Haitian asylum applications, then 
the most controversial portion of its caseload. If UNHCR disagreed 
with a draft State advice letter, it raised its concerns with BHRHA 



'** See General Accounting Ofhce, supra note 150, at 25; P. Fagen, supra note 
141, at 54-55. The problem is by no means confined to asylum cases. Other studies 
have noted the general inability of the INS system to secure actual deportations. See 
E. Harwood, supra note 66, at 41-46. In 1986, INS also changed the regulations to 
eliminate the 72-hour advance notice to surrender for deportation given to aliens 
already subject to a final deportation order, in part because an INS study found that 
more than 76 percent of recipients absconded after receiving such letters (often 
known familiarly as "bag and baggage" letters or, even more familiarly, as "run 
letters"). 5*^ 51 Fed. Reg. 3471 (1986) (to be codified at 8 C.F.R. § 243.3) (proposed 
Jan. 28. 1986); id. at 23,041-42 (1986) (codified at 8 C.F.R. § 243.3 (1989)). 

•89 5^^ generally E. Harwood, supra note 66, at 30, 122-30, 184 (discussing 
enforcement practices). 

•90 See id. at 41-46 (reporting the same phenomenon). Harwood recounts the 
story of one alien who showed up two days before the date shown on her "bag and 
baggage" letter for deportation to El Salvador. The INS office made her wait. One 
officer explained: "What they have to realize is that deportation is a privilege, not a 
right." Id. at 46. 



306 DAVID A. MARTIN 

and a negotiating process ensued before the hnal letter was sent. 
Sometimes the Bureau persuaded UNHCR to change its views. In all 
other cases of initial disagreement, the State Department accepted 
the UNHCR position. '^^ 

Although there have been frequent calls for expanding this prac- 
tice to cover all asylum cases, during the 1980s the trend went in the 
opposite direction. The Reagan administration took a more skepti- 
cal stance toward international organizations (to put it mildly), and 
UNHCR's access declined. *^^ Although the Washingfton office 
retains some contact with State and INS to communicate its general 
views on, for example, proposed regulations, its role as systematic 
reviewer of individual cases has long since ended. It now performs 
three other functions: (1) helping asylum seekers, on occasion, to 
locate pro bono counsel; (2) filing amicus curiae briefs in cases (usu- 
ally class actions) likely to have wide impact on asylum processing; 
and (3) sending a letter to applicant's counsel expressing UNHCR 
views on the particular application. UNHCR receives far more 
requests for the last service than it can possibly meet. It tries to be 
selective and write such a letter only in strong cases, and only at the 
stage when the matter is already before an immigration judge. It is 
then up to counsel to introduce the letter in an appropriate manner 
into the immigration court proceedings.*®' 

8. The August 1987 Proposed Regulations 

Throughout the 1980s, INS has been considering further 
reforms to the asylum process to make it more expeditious and effec- 
tive. The most thorough effort at crafting reforms was built upon an 
internal study carried out by Richard Spurlock, a retired district 
director hired as a consultant for these purposes in 1985. Because of 
the sensitivity of the issues, the potential costs, and the multiple 
agencies involved, each with its own particular angle of vision and 



'®' See U.S. Refugee Programs: Hearing before the Sen. Comm. on the fxidiciary, 96lh 
Cong., 2d Sess. 15 (1980) (testimony of Cyrus Vance, Secreury of State). Secreury 
Vance expressed satisfaction with this arrangement and stated: "[t]his doublecheck 
which we have developed by working with the U.N. High Commissioner is a sensible 
and wise way of checking our standards and seeing that they are being fairly applied." 
Id. 

'^2 5^^ Burke, supra note 163, at 325 (presenting argument of one official of the 
Reagan Administration against UNHCR participation in asylum adjudication because 
determining who may stay "is a fundamental attribute to sovereignty"). 

1^' An example of such a letter may be found in Dwomoh v. Sava, 696 F. Supp. 
970. 979 (S.D.N.Y. 1988). 



REFORMING ASYLUM ADJUDICATION 307 

bureaucratic turf to protect, the proposals did not result in formally 
promulgated draft regulations until August 1987.'^ 

Those proposed rules ranged widely and contained some 
improvements broadly supported. But the central change touched 
off a storm of controversy. The regulations proposed to establish a 
new corps of asylum adjudicators in INS, responsible to the Central 
Office's Office of Refugee Asylum and Parole, which would consider 
all asylum issues, no matter at what stage of the proceedings the asy- 
lum claim was filed.*®* The immigration judges would have been 
removed from asylum issues altogether, thus ending the provision of 
two possible rounds of de novo consideration. 

If the alien were already in deportation or exclusion proceed- 
ings at the time of the asylum claim, those proceedings were to be 
adjourned to permit consideration of the case by the new adjudica- 
tors. The State Department was to receive a copy of all asylum appli- 
cations and retained the option of communicating its views to the 
adjudicators, but the proposal probably would have led to a reduced 
role for the Department. If asylum was denied, the matter would 
return to the immigfration court for consideration of any other 
defenses, and for issuance of a final deportation or exclusion order. 
Under the proposed regulations, the immigration judge had no 
authority to reconsider the asylum claim. The alien could still appeal 
to the BIA, however, and the Board retained authority to review the 
adjudicator's asylum decision. 

These regulations evoked a strong reaction from refugee advo- 
cates. Although some had earlier expressed openness to the idea of 
a single corps of expert adjudicators, all were deeply concerned 
about the adjudicators' lack of independence under the August draft. 
The asylum office was to remain in INS, instead of being moved to 
EOIR or a wholly new independent agency. More over, initial indi- 
cations about staffing and training held litde promise of significant 
upgrading in the quality of personnel or procedures over that 
already in the district offices.*^ 

Although some comments filed in response to the rulemaking 
suggested measures that would address these concerns direcdy, 

>»4 See 52 Fed. Reg. 32.552 (to be codified 8 C.F.R. pts. 3. 208. 236. 242. 253) 
(proposed Aug. 28. 1987). 

•95 5^^ td. Aldiough the regulations did not so slate. INS initially envisioned 
sutioning these officers in seven or eight cities throughout the country, with some 
provision for "circuit riding" to hear claims lodged in more remote locations. 

•9« See Helton, The Proposed Asylum Rules: An Analysts, 64 Interpreter Releases 
1070. 1073-74 (1987). 



308 DAVID A. MARTIN 

while retaining the basic idea of a single round of adjudication 
before a specialized set of adjudicators, NGO opposition soon 
focused with vigor on one particular cure: reinstatement of the 
opportunity for de novo consideration by the immigration judges. 
The campaign was so intense that the matter moved directly to the 
desk of the Deputy Attorney General, who decided to accede to the 
NGO position. On the day the comment period closed, the Justice 
Department told the press that new regulations would issue reinstat- 
ing the judges in the process. '^^ Finally, in April 1988, new pro- 
posed regulations appeared, implementing this decision, but 
continuing with plans for a centralized corps of adjudicators to 
replace the district office examiners.^®® To date, continuing internal 
disputes have prevented final promulgation.*^ 

C. Evaluation 

Almost no one regards the current asylum adjudication system 
as an effective and efficient scheme for deciding on what Judge Ken- 
neth Starr (now Solicitor General) has called "this most sensitive of 
human claims in the international community."***^ If accuracy, 
speed, and fairness are the key objectives in asylum adjudications, 
the current system achieves them in only a small portion of cases. 

1. Speed and Fairness 

a. Two Bites at the Apple 

Despite nominal agreement on all sides that expeditious pro- 
ceedings are needed, the current system rarely achieves prompt 
finality. The most obvious culprit is the wasteful provision of two 
venues for de novo rulings in asylum— exacdy the problem targeted 
in the August 1987 proposed rules. ^°' Those rules, of course, met 

'9' See Arocha, Political Asylum Revision Dropped, Wash. Post., Oct. 30, 1987, at 
A23, col. 1. This was followed by a formal announcement to this effect in the Federal 
Register in December. See 52 Fed. Reg. 46,776 (to be codified 8 C.F.R. pts. 3, 208, 
336, 242, 253) (proposed Dec. 10, 1987). 

'98 See 53 Fed. Reg. 11.300 (to be codified 8 C.F.R. pts. 3, 208, 236, 242. 253) 
(proposed April 6, 1988); Helton, Asylum Rules Revtsiled: An Analysis, 65 Interpreter 
Releases 367 (1988). 

'^ See, e.g., 66 Interpreter Releases 3 (1989) (suggestion by INS 
Commissioner that asylum adjudication function should remain in the district offices 
as before). 

200 Reyes-Arias v. INS, 866 F.2d 500, 504 (D.C. Cir. 1989). 

201 See Kurzban, Restructunng the Asylum Process, 19 San Diego L. Rev. 91, 98. 
111-12 (1981). 



REFORMING ASYLUM ADJUDICATION 309 

with such strong opposition that the Justice Department beat a hasty 
retreat and reinstated the role of the immigration judges in revised 
proposed regulations issued in April 1988. But the nature of the 
objections, the proposed remedy of the NGOs, and the Administra- 
tion's ultimate response merit further inquiry. 

Most of the opposition to the August 1987 proposed rules 
derived not from a belief that the immigration court provides the 
ideal setting for consideration of asylum claims. Indeed, the judges 
are often criticized in other venues, on a variety of grounds, by the 
same people who attacked the 1987 proposal. The opposition 
derived instead from concern about the quality of decision-making 
that could be expected under the precise form of unification that was 
proposed. The centralized corps of asylum adjudicators, who would 
have become the sole arbiters, would not likely have been much dif- 
ferent, in training, background, or oudook, from the current examin- 
ers who make the decisions in the district offices. Opponents of the 
new regulations were able to collect affidavits with numerous stories 
of brusqueness, mishandling, errors, and apparent bias on the part 
of some of those officials.^®^ 

Early internal versions of what became the 1987 proposed regu- 
lations had considered assigning the newly centralized adjudication 
function to a different set of officials — attorneys at a higher civil ser- 
vice rank. But the Administration ultimately chose instead a version 
that would keep the position one for journeyman immigration exam- 
iners, and OMB initially assigned to the new positions, a relatively 
low grade (relative, that is, to the magnitude and challenge of the 
adjudication required) of GS-11 and GS-12 for supervisors. The 
administration decided, in short, to attempt reform on the cheap, by 
shifting boxes on the organization charts rather than investing ade- 
quately in a new system and new personnel that might break through 
the established, and destructive, patterns of polarization and distrust 
that have paralyzed effective reform for years. The vigorous reaction 
from the NGO community should not have been a surprise, given 
that the adversarial forum was being replaced with such a stingy 
alternative. 

Some old government hands in the immigration field, familiar 
with the shifting patterns of NGO advocacy over the last fifteen 

202 5^ supra note \AA; see abo Note, Asylum Adjudtcation: No Place for the INS, 20 
CoLUM. Hum. Rts. L. Rev. 129. 143-44 &: n.l04 (1988) (arguing that "[tlhe 
impartiality of the INS as an adjudicatory authority is compromised by the emphasis 
on enforcement which pervades the entire agency and by the direct effect this 
mentality has on decision-making"). 



310 DAVID A. MARTIN 

years — sometimes favoring nonadversarial procedures, sometimes 
treating the immigration judges as indispensable— conclude that the 
position of the NGOs is always and only a tactical one, meant to pre- 
serve "two bites at the apple" whatever the proposal on the table. 
But to be fair, there is no inherent inconsistency in the advocates' 
position. If speed were no concern, it might well be that the best 
possible system does involve two rounds of de novo consideration in 
different institutional settings: first a nonadversarial hearing to fos- 
ter responses from hesitant or fearful applicants, followed by an 
adversarial, trial-type hearing that we traditionally identify as the 
best way to honor due process when the stakes for the individual are 
high.^^' Arthur Helton, a leading figure in the asylum debates, 
stated forcefully during her interview for this study that the fight 
over the 1987 regulations has left the NGO community deeply com- 
mitted to "bifurcated proceedings," that is, to a system that allows 
two separate forums for initial, de novo consideration.^*** NGOs will 
probably resist stoutly any departure from the victory they feel they 
justifiably won in October 1987. 

But here is the rub: Speed is a concern, a vital concern. It must 
come to be seen as such by the NGOs as well as government officials, 
including high level Justice Department officials who step in on 
immigration matters only when political controversy bums high. 
Without speedy denials, the system will either attract large numbers 
of marginsd claimants or else force resort to other cosdy and trouble- 
some deterrents which indiscriminately burden genuine refugees. 
Restoring the "two bites" system thus implicated far greater costs 
than were appreciated in October 1987, by either the NGOs or the 
Justice Department. 

b. Administrative and Judicial Review 

Delay also derives from the clogged dockets of the immigration 
courts and from backlogs at the BIA. Shortening those delays 
requires additional resources, rather than major institutional rede- 
sign,^°* and remains within the general control of the Justice Depart- 
ment, provided, of course, it receives adequate appropriations. 

203 See VcrkuU, A Study of Immigration Procedures, SI UCLA L. Rev. 1141. 1145 
(1984) (identifying Goldberg v. Kelly, 397 U.S. 254 (1970), which imposed extensive 
requirements for trial-type hearings, as the "high watermark of procedural due 
process in the administrative setting"). 

20< Interview with Arthur Helton, in New York City (Nov. 15, 1988). 

^^ The recent rules changes allowing the BIA to sit in panels of three should 
help reduce that backlog. See 8 C.F.R. i 3.1(a)(1) (1989). 



REFORMING ASYLUM ADJUDICATION 3 1 1 

(This is a powerful proviso, of course, in these days of Washington 
lip-reading.) One particularly important element in the delay, how- 
ever, derives from the lengthy period of time required to obtain tran- 
scripts of immigration court proceedings. That period can now run 
eight months or more, although cases involving detained aliens are 
given priority and can be processed within a matter of weeks. If 
expeditious final rulings are to be received, EOIR must either 
arrange for a quicker turnaround of transcripts or else experiment 
with alternatives to full transcription of each hearing.^**^ 

Major delay also potentially derives from the provisions for judi- 
cial review, although this does not appear to affect a significant pro- 
portion of cases at present. Several refugee attorneys stressed 
during interviews that judicial review is sought only if the case 
appears particularly strong — in part because of the pro bono nature 
of the representation, and in part because such attorneys worry 
about developing bad law that would serve to undermine stronger 
cases later. This practice thus contrasts importandy with the use of 
other stages of the process, because immigration attorneys, I was 
told, consider renewal of asylum cases in immigration court and 
administrative appeals to the BIA to be routine steps in all but the 
most farfetched cases. 

Changes to the judicial review provisions would therefore not 
appear to be warranted at this time. If these patterns change, and 
judicial review someday comes to cause delay in a far higher percent- 
age of cases, then reform could be considered at that point. Statu- 
tory changes would then be needed. Other nations confronting the 
problem of judicial delay have tried two approaches: (1) by speedily 
identifying a class of applications adjudged "manifesdy unfounded" 
and stricdy limiting judicial review for that class; and (2) more ambi- 
tiously, by limiting the scope of judicial review in virtually all asylum 
cases to a summary proceeding that is highly deferential to the 
administrative outcome, but affords some opportunity for judicial 
correction of gross error or abuse. TTiese approaches will receive 
greater, albeit preliminary, attention below in connection with pro- 
posed reforms. 

c. Is Delay Really a Problem? 

Before turning to the alternative approaches, however, one fur- 

206 A Canadian study has recommended the use of videotapes, with briefs citing 
the location on the tape where crucial matters appear. See E. Ratushny, supra note 
64. at 57-58. 



312 DAVID A. MARTIN 

ther set of objections to the above evaluation should be aired. Refu- 
gee advocates sometimes argue that the problems of administrative 
and judicial delay are exaggerated. They agree that the current sys- 
tem allows, at least theoretically, for six layers of consideration 
(three administrative and three judicial) in exclusion cases, and five 
in deportation cases, but they point to the absence of statistics show- 
ing that many asylum seekers actually avail themselves of all these 

• ■ 907 

opportunities. 

It is certainly true that the case for administrative simplification 
cannot be made convincingly on the basis of currently available sta- 
tistics. EOIR maintains and releases only limited statistics on asylum 
caseloads, and it is impossible to tell how many of its asylum cases 
represent renewals of applications initially rejected in the district 
offices.^**® The GAO study based on 1984 applications found that 
only seven percent of applicants denied in the distria office renewed 
their claims before the immigration judges.***® But this statistic is 
suspiciously low, and may be attributable to the fact that time limita- 
tions on the study precluded the GAO from tracking all of the cases 
initially denied through to the applicants' actual removal from the 
country or to some other resolution of their status. It would be 
entirely possible that many aliens involved simply were not 
processed for deportation until after the study period ended, partic- 
ularly because the study found a low INS priority on initiating such 
deportation cases.**® Everyone interviewed for the present study — 
including immigration lawyers — thought that the seven percent fig- 
ure was too low, although most placed their rough guesses of actual 
renewals in the 20-30 percent range, an estimate still not terribly 
high. Similarly, judicial statistics (ag^in limited) do not show massive 
court litigation over asylum.*** 

These statistics, however, almost certainly tell a dated story. 
Before the full implementation, in 1988, of sanctions on employers 



207 See, KuTzban. si^na note 201. at 96-97. 

*08 It may be that the more complete EOIR statistics being released as a result 
of a policy change in early 1989 will someday provide greater insight into these 
issues. See supra note 160. 

209 General Accountinc OrncE, si^tra note 150, at 20, 33. At the time of the 
study, 77 percent of the claims had been filed only in district offices and 16 percent 
only in immigration court. 

210 See id. iX. 27-29 (describing methodology), id. at 38 (showing that for 81 
percent of the cases INS had taken no "deportation action" by the close of the study 
period). 

2" Set supra Table V following note 187. 



REFORMING ASYLUM ADJUDICATION 313 

who hire undocumented aliens,^ *^ persons denied asylum in the dis- 
trict office had little incentive to pursue matters further. Most could 
probably find work and enjoy the benefits of a "de facto asylum" that 
carried few risks, despite their undocumented status.^** Such aliens 
had no reason to rush further review, because they always retained 
the option of renewing the asylum application in immigration court 
if and when the INS caught up with them and initiated deportation 
proceedings. Moreover, because of enforcement priorities, deporta- 
tion has not been an imminent threat. 

Since Congress's adoption of employer sanctions, all of this has 
changed, although it is still too early for reliable statistics. The 
"walk-in" rate is up considerably in the district offices, and evidence 
suggests that, unlike in earlier years, many people now choose to file 
affirmatively for asylum in order to receive work authorization.^** 
Once denied, they then have incentives to press INS to initiate 
deportation. In a striking role reversal, the immigration bar and ref- 
ugee support groups have been urging INS to hastm arrangements 
for the issuance of orders to show cause in these circumstances, in 
order to assure an early forum for rearguing the asylum claim and, 
by no means incidentally, to renew work authorizations.^'* Past sta- 
tistics thus furnish no reliable guide as to the magnitude of the delay 
problem under the present multilayered administrative system, given 
the new employer sanctions. 

d. Rights to Counsel 

One further element of possible delay lurks in the current sys- 
tem, owing to a major feature meant to enhance fairness to the appli- 
cant. Statute and regulation provide a right to counsel in the 
immigration court proceeding, although at no expense to the gov- 

212 The employer sanctions scheme, INA § 274A, 8 U.S.C. § 1324a (1988), was 
enacted as part of the Immigration Reform and Control Act of 1986, Pub. L. No. 99- 
603, 100 Sut. 3359. 

213 See generaUy Martin, The End of De Facto Asylum, supra note 105, at 163-64 
(1987-88). 

21* S« B. Frelick, The Back of the Hand: Bias and Restrictionism Towards 
Central American Asylum Seekers in North America 17 (U.S. Comm. for Refugees 
1988). 

215 See Memorandum from Ignatius Bau and Robert Rubin (July 13, 1987), 
reprinUd in 64 Interpreter Reuj^ses app. II at 882, 886 (1987); see also INS Advises on 
Work Authorizations for Denied Asylum Applicants, 65 Interpreter Reixases 718 (1988). 
Similar incentives may someday operate at the judicial review suge, at least as long as 
the work authorization regulation remains unchanged, but that appears a more 
remote prospect, given the reluctance of most experienced asylum attorneys to take a 
weak case to court. See supra text following note 206. 



314 DAVID A. MARTIN 

emment. They also mandate the provision of a list of pro bono 
counsel available in the area.^*® When asylum claims rise numeri- 
cally, the limited numbers of pro bono attorneys are easily 
swamped.^*' This situation places immigration authorities in a diffi- 
cult position as they try to keep pace with rising intake — a perfecdy 
legitimate and praiseworthy bureaucratic objective. Inevitably, they 
have incentives to press for waivers of counsel or else to deny 
repeated continuances requested because of free counsel's limited 
availability. Recent court decisions, however, are imposing increas- 
ingly demanding requirements to assure a knowing waiver of counsel 
rights. Concomitandy, courts are increasingly insistent that the 
immigration judges allow continuances until pro bono counsel 
becomes available.^*® Obviously, in times of major influx, or in areas 
where counsel is limited, this stance can cause backlogs to increase 
substantially. 

Court reversals for failure to honor these (qualified) counsel 
rights are particularly demoralizing to the system, for such reversals 
inevitably come months or years after the initial proceedings and all 
later phases of administrative consideration, at a time when the 
deportation or exclusion order is administratively final. Moreover, 
such reversals plainly require complete rehearing in the immigration 
court, possibly followed again by administrative and judicial 
review.^*® If the system is to be prepared reliably for speedy deter- 
minations despite a fluctuating caseload, the problem of counsel 
must be solved, either by wider provision of government-paid coun- 
sel^^^ or by assuring fairness to the applicant even in the absence of 
counsel. 



816 See INA H 242(b). 292. 8 U.S.C. §fi 1252(b). 1362 (1988); 8 C.F.R. pu. 292. 
292a (1989). 

217 For an example of these complications, see Maldonado-Perez v. INS, 865 
F.2d 328, 329-32 (D.C. Cir. 1989). 

218 See. e.g.. Baires v. INS, 856 F.2d 89. 92-93 (9th Cir. 1988): Castro O'Ryan v. 
INS. 847 F.2d 1307. 1312-13 (9th Cir. 1988); Haitian Refugee Center v. Smith. 676 
F.2d 1023, 1031 (5th Cir. 1982). Some cases are more tolerant of the agencies' 
efforts to process cases in this manner. See, e.g., Committee of Central American 
Refugees v. INS. 682 F. Supp. 1055. 1065 (N.D. Cal. 1988); Vides-Vidcs v. INS, 783 
F.2d 1463. 1470 (9th Cir. 1986). 

2 '9 Class actions challenging broad features of asylum processing can likewise 
lead to this result. See supra note 3. 

220 At least one commenutor regards the Equal Access to Justice Act (EAJA) as 
a possible solution to these problems. See Note, Applying the Equal Access to Justice Act to 
Asylum Hearings, 97 Yale L.J. 1459 (1988). But EAJA is at best a half-measure that 
affords only marginal relief, because attorneys and agencies cannot know up front 
whether or not they will actually be compensated from the public treasury. It may be 
years before the fee question is settled. At best EAJA may attract a few more 



REFORMING ASYLUM ADJUDICATION 315 

The reforms proposed below center on a nonadversarial model 
that could provide a full and fair opportunity to present an asylum 
claim, even if the individual is unrepresented. Although efforts 
would be made to accommodate counsel's schedule, the proposed 
system is designed to proceed with fairness on a fairly prompt time- 
table, even if the applicant expresses a desire for pro bono counsel 
but insufficient numbers of counsel are available to meet the 
demand. It must be acknowledged that such a proposal will be 
highly controversial. Due process, at least when the stakes are high, 
is often closely associated with adversarial trial-type proceedings, 
which usually require professional counsel. 

If the nonadversarial model is not accepted as the way of provid- 
ing fairness to unrepresented asylum applicants, some carefully 
crafted alternative is imperative. It is essential to maintain the sys- 
tem's capacity for expeditious processing, without the risk that a 
court will send the case back to square one months or years later. 
The best alternative is probably for Congress to amend the statutes 
to eliminate the ban on government-paid counsel — not for all immi- 
gration matters but only for nonfrivolous asylum cases. 

Because of unpredictably fluctuating caseloads, it would proba- 
bly be best, if such a course is chosen, to follow a public defender 
model, assigning the representation responsibility to a permanent 
staff of government-paid attomeys.^^' Such a staff would provide 
other advantages as well, because of the expertise they could develop 
concerning country conditions over the course of litigating numer- 
ous cases. Naturally, applicants could still retain private counsel or 
engage unaffiliated pro bono attorneys when available, but the pace 
of proceedings would no longer be so dependent on private char- 
ity.^^^ Obviously, the cost of such a system would be substantial, and 



attorneys into initially pro bono representation, and groups currently providing 
assistance may be able to expand their staffs using EAJA awards from past cases. But 
unless the expansion far exceeds expectation, pro bono attorneys are still unlikely to 
be able to keep pace with new arrivals in times of large influx. Either delays will 
continue or many applicants will remain without representation — again risking 
judicial remand many months hence. 

221 Arrangements could be made for adequate insulation of these officers to 
assure an independent litigation posture, or perhaps the function could be 
performed under contract with a nongovernmental organization that would agree to 
maintain an adequate legal staff at the necessary locations, paid from government 
grants. Briuin has arrangements of roughly this sort with the United Kingdom 
Immigrants Advisory Service (UKIAS), although not all of UKlAS's representatives 
are lawyers. See generally European Consultation on Refugees and Exiles, Asylum 
IN Europe: A Handbook for Agencies Assisting Refugees 371 (3d ed. 1983). 

222 As an added benefit. NGOs might then find it easier to target their own 



316 DAVID A. MARTIN 

the visibility of an appropriation for such purposes, at a time of 
extreme budgetary stringency, makes this course politically unlikely. 
But the current statutes (allowing counsel only "at no expense to the 
government") provide only illusory cost savings. Although no 
money goes directly to the applicant's counsel, the government 
incurs substantial expenses, primarily from detention prolonged by 
the period necessary to wait for the pro bono counsel to become 
available. Other indirect costs are harder to quantify but probably 
more substantial — for example, the burdens on local services caused 
by massive influxes of asylum seekers. ^^^ A speedy system is impera- 
tive if such burdens and costs are to be reduced. 

2. Accuracy 

a. Diffusion of Responsibility 

The current system fails to focus responsibility for this difficult 
and challenging decisionmaking on one set of officials. It thus 
enhances the risk of improper denials of asylum, even if all officials 
act in good faith. Over the years, INS spokespersons have some- 
times responded to complaints about asylum denials by pointing out 
that all cases are checked yfirh the State Department, which has 
expertise concerning these matters, and that INS almost always fol- 
lows the Department's lead.^^* At the same time, however. State 
Department officials have often felt that it was INS or the immigra- 
tion judges who really performed the important part of asylum deci- 
sionmaking. Department officials, after all, see only the printed 
page. Much of the adjudication must turn on credibility judg- 
ments—clearly a task principally for Department of Justice adjudica- 
tors, who see the applicants in person and can test the stories 
through direct questioning.^^* The system thus courts the risk that a 
negative State Department opinion will induce some relaxation in 
Justice Department care in examining individual cases. Yet that 

scarce resources on cases they regard as most deserving, or as critical for establishing 
an important Victual or legal precedent. 

22S See Appendix (for rough cost estimates). 

224 See generally Asylum Adjudications, supra note 142, at iii, 61-64. Similarly, a 
1983 study of asylum processing in New York noted "a certain discomfon with 
asylum cases" among immigration judges that led to heavy reliance on State 
Department views and a some disavowal of respK>nsibility. See P. Fagen, supra note 
141. at 16. 

225 See AsYLin^ Adjudications, supra note 142, at 60. I also encountered this 
attitude with some frequency during my own service in BHRHA (known simply as 
HA in State Department lingo) from 1978 to 1980. 



REFORMING ASYLUM ADJUDICATION 317 

opinion may have been issued in recognition that a cold record is not 
fiilly revealing, and in anticipation that the alien will have another 
chance to bolster her case before an adjudicator she will see in 
person. 

It is not an easy thing to send a person back to a land where she 
claims she faces persecution. Unsurprisingly, officials may therefore 
seek at times to minimize their own role in such results. But arrange- 
ments that unintentionally help to meet that psychological need may 
entail systemic costs. A system that provides undue comfort in going 
along with negative results may fail to create adequate incentives for 
the care needed to spot the truly meritorious case.^^® 

The reduced role for the State Department under the new 
sticker system should ameliorate this problem. A more complete 
focusing of the decision on a single set of adjudicators would provide 
even greater assurance. It would also give applicants the opportu- 
nity to make their cases in person to the official who will be responsi- 
ble for all phases of the decision. 

b. Political Bias 

For decades the asylum adjudication system has been attacked 
for the bias of its results. ^^' No completely reliable scientific test of 
these claims is possible, and a GAO study chartered to discover 
whether asylum applications were g^ven neutral consideration could 
only conclude that the matter was "uncertain."^^® Nevertheless, 
even a quick glance at the statistics in Table II raises serious ques- 
tions about the high grant rates for applicants from communist coun- 
tries (particularly Poland, where political activity became much freer 
after the lifting of martial law in 1984), and the strikingly low rates 
for El Salvador and Guatemala. Moreover, testimony about bias 
comes in highly persuasive form from published accounts by INS 
insiders as well as from INS's critics. ^^^ 



226 See Anker &: Posner. nipra note 31, at 76; Aleinikoff, supra note 89, at 193-94. 

227 See, e.g.. G. Loescher & J. Scanlan, supra note 2, at 170-219; N. L. Zucker &: 
N. F. Zucker, supra note 61, at 142-76; Helton, Political Asylum Under the 1980 Refugee 
Act: An UnfulfilUd Promise, 17 U. Mich. J.L. Ref. 243. 260-62 (1984); Kurzban, A 
Critical Analysis of Refugee Law, 36 U. Miami L. Rev. 865, 876-78 (1982); Note, Political 
Bias m Uniud States Refugee Policy Since the Refugee Act of 1980, 1 Geo. Immicr. L.J. 495 
(1986) [hereinafter Political Bias]; Note, Behind the Paper Curtain: Asylum Policy versus 
Asylum Practue. 7 N.Y.U. Rev. L. & See. Change 107, 107-09. 124-26 (1978). 

228 5^^ General Accounting Office, supra note 150. 

229 See Asylum Adjudications, supra note 142, at 59 n.*; Meissner, supra note 2, 
at 57, 63 (describing the pressures that skewed decisionmaking, with special attention 
to Poland and El Salvador; the author was Acting Commissioner of Immigration from 



318 DAVID A. MARTIN 

Much of the outsiders' criticism blames bias on the role of the 
State Department (recently modified) in providing advisory letters 
on every case considered in the district offices and immigration 
courts. ^'° Clearly that practice provides an opportunity for diplo- 
matic considerations to intrude on what the statute ordains should 
be neutral, case-by-case decisionmaking.^** For this reason, it would 
be far better to remove the Department from any substantial role in 
the decisionmaking system.^'^ No matter how conscientious the 
State Department may be in performing this function, the aura of 
distortion is bound to linger. Moreover, the Department itself would 
benefit from such separation. When the home-country government 
is angered by an asylum grant, its ire can be more easily deflected by 
our diplomats if the Department can credibly state that it had no role 
in the decision.^" 



1981-83 and Executive Associate Commissioner fix>m 1983-86). In unguarded 
moments, some Reagan Administration spokespersons also revealed that they 
regarded asylum adjudications as inherently political. For example, in arguing 
(unsuccessfully) for extradition reform legislation that would have transferred from 
the coiuts to the State Department the authority to decide whether a particular 
offense was "political" and hence non-extradiuble, the Deputy Legal Adviser stated: 
"[A] decision on the 'political offense' exception is (as the name suggests) inescapably 
political in nature, and inextricably intertwined with the conduct of foreign relations. 
It is an issue best left to the Executive branch to decide— mucA as the decision to offer 
political asylum is an executive decision. " Extradition Reform Act of 1981: Hearirtgs on H.R. 
5227 Before the Subcomm. on Crime of the House Comm. on the Judiciary, 97th Cong.. 2d 
Sess. 32 (1982) (prepared statement of Daniel W. McGovem) (emphasis added). 

^^ See, e.g., Posner, Comments and Recommendations on Proposed Reforms to United 
States Immigration Policy, 36 U. Miami L. Rev. 883. 887 (1982) ("[0]nc of the most 
troubling problems with the current immigration system is the Sute Department's 
involvement in the decisionmaking process."); Political Bias, supra note 227, at 537 
("Tlie chief problem concerning residual political bias in asylum adjudications is the 
critical role that the State Department plays in assessing asylum claims."). 

2S> See, e.g, Zamora v. INS. 534 F.2d 1055. 1062 (2d Cir. 1976) (noting "some 
likelihood of the Department's tempering the wind in comments concerning the 
internal affairs of a foreign nation"); Kasravi v. INS, 400 F.2d 675. 677 n.l (9th Cir. 
1968) ("A frank, but ofRcial, discussion of the political shortcomings of a friendly 
nation is not always compatible with the high duty to maintain advantageous 
diplomatic relations . . . ."); Aleinikoff, supra note 89, at 194, 234-35. 

252 por an argument that the State should reuin its role, see Burke, supra note 
163. ai 325. 

2" Some system should still be worked out for regularly notifying the State 
Department of cases received, to help prepare it for cases that will spark political 
controversy with the home government, to assure identification of any asylum 
seekers who may prove to be defectors with sensitive information, and perhaps to 
allow for the Department or the intelligence agencies to introduce confidential 
information bearing on the claim, pursuant to procedures permitting use of such 
information in limited circumsunces. See. e.g, 8 C.F.R. § 208.10(c), (d) (1989) 
(setting forth procedures for disclosure of non-record evidence). Elach eventuality is 
rare in asylum cases. 



REFORMING ASYLUM ADJUDICATION 319 

Removing the State Department from asylum processing has 
been advocated for years. But many critics fail to think through care- 
fully the continuing risks of political bias even then, unless other 
important changes are made as well. The "coast of Bohemia" prob- 
lem would still be present.^*^ Indeed, it would probably be aggra- 
vated. As Lippmann observed in her classic study, those who know 
less about the realities of an issue or a far-away land are more likely 
to rely on the "pictures in their heads" to cope with a challenging 
and complex mass of data.^'* Without State Department advice, 
nonexpert adjudicators will be confronted with just such a dilemma. 

A 1986 episode, widely reported in the newspapers, reflects the 
impact of such stereotyped thinking. The district director in Miami 
decided in April of that year to end all returns of aliens to Nicaragua 
from her district, because of concern that they might be persecuted 
by the Sandinista government. She was quoted as saying: "I would 
personally — ^not just as a Government official, but personally — ^have 
trouble sending people from a Communist country back to that 
country."^'® Although this position was at odds with State Depart- 
ment information (as reflected in the annual country reports and in 
the fact that numerous advisory letters for Nicaraguans were still 
negative), and contravened Congress's explicit decision in 1980 to 
adopt a neutral standard to replace former provisions that expressly 
favored refrigees from Communist countries, the Justice Department 
apparendy made no effort to discipline the district director or other- 
wise bring her actions back into line.^'^ 

The phenomenon is not confined to the district offices. Some 
immigration judges also volunteered to me during office interviews 
that they considered a certain country (usually a Communist coun- 
try) too dangerous for return, although none offered very complete 
or convincing reasons for this judgment. Certainly they did not 
claim that it was based on State or Justice Department policy or on 

254 Even today, the State Department's advice is not always followed. Some 
impressionistic evidence, however, suggests that the failure to follow the advice has 
its own bias that compounds the favoritism for those who flee Communist countries. 
See T. Aleinikoff & D. Martin, st^a note 3, at 705. 

255 w. Lippmann, supra note 81, at 30-49. 

256 Pear, Key Federal Aide Refuses to Deport any Nicaraguans. N.Y. Times. Apr. 17, 
1986, at A 1, col. 1. 

237 5^^ ^ ; Dreifus, supra note 72, at 35. It appears likely that this action, along 
with other Department leniency toward Nicaraguans over the last two years, became 
well-known back in Nicaragua and played a role in eventually encouraging large 
numbers of people there to think about migration to the United States. The fruits of 
that encouragement were felt in the Harlingen district, at a rate of several thousand 
marginal applications for asylum each month during the %^nter of 1988-89. 



320 DAVID A. MARTIN 

country guidelines or across-the-board authoritative findings. It was 
simply based on their views about current world conditions. In the 
context of the conversations, I would not regard these comments as 
deliberate bias; they were meant as sincere efforts to implement the 
statute's commands in light of that person's understanding of home- 
country conditions. But especially when matched up with competing 
stereotypes that may lead to great skepticism of claims from Central 
Americans other than Nicaraguans, the potential for inaccurate 
results and for unfairness is manifest.^'® 

Most immigration judges, it should be stressed, strive conscien- 
tiously to apply the legal standards fairly based on the records before 
them. But the potential for improvement remains. Some better way 
should be found to correct as much as possible for the unintended 
bias that derives from the adjudicators' inevitable creation of internal 
maps, particularly if the State Department is to be removed fi:*om 
routine involvement. Systematic effort should be undertaken to 
replace stereotypes with detailed and accurate information, helpfully 
digested. The "coast of Bohemia" problem can never be eliminated, 
but it can be minimized. 

c. Inadequate Use of Existing Expertise 

The immigration judges* role is firmly anchored in the adver- 
sarial model; by and large, they are expected to remain as passive 
arbiters ruling on records developed by the pjuties.^'® Initiative by 
the judges to learn more about country conditions is not officially 
encouraged. EOIR Chairman Milhollan has specifically rejected an 
AILA suggestion that judges should receive more training on coun- 
try conditions, and reaffirmed the traditional adversarial model.^*° 

2'B See P. Fagen, supra note 141, at 21 (finding that presumptions based on 
national origins distorted asylum adjudications). 

239 In this respect, of course, EOIR is simply trying conscientiously to move 
away from a problematic past, when immigration procedures were harshly criticized 
for their inquisitorial character and for the mixture of enforcement and adjudicative 
roles for the special inquiry officers. See T. Aleinikoff & D. Martin, supra note 3, at 
87-91. This effort is praiseworthy in most immigration-law settings, but certain 
elements fit uneasily in the asylum proceeding. Most troublesome is the assumption 
that the judge is a kind of blank slate at the beginning of each new case. See generally 
S. Landsman, Readings on Adversarial Justice: The American Approach to 
Adjudication 2-S (1988) (identifying a "neutral and passive decision-maker" as a key 
element of the adversarial model). This specialized kind of adjudication, unlike most 
other matters within the immigration courts' jurisdiction, cannot be p>erformed well 
unless the adjudicator brings to the case cumulative expertise concerning country 
conditions. See E. Ratushny, supra note 64. at 15-18, 51. 

240 fg^w immigration judges, however, are given specific training on asylum 



I 



REFORMING ASYLUM ADJUDICATION 321 

She stressed "that it was the attorney's responsibility to offer 
whatever evidence she or she deems appropriate to meet the alien's 
burden of proof in asylum cases. "^'** 

Officially, therefore, immigration judges and the Board of Immi- 
gration Appeals must decide asylum cases based only on the record 
created in the specific case.^**^ In reality, particularly when numer- 
ous cases are received from the country involved, the judges and the 
Board cannot help but remember, and, to some extent, use informa- 
tion learned in other cases. In fact, some immigration judges have 
become quite knowledgeable about conditions in those countries, 
especially in Central America, whose nationals account for a high 
percentage of asylum claims. Indeed, this is a praiseworthy practice 
that probably helps improve the quality of decisionmaking, even 
though it violates the formal requisites of the adversarial model. 

One immigration judge I interviewed had in her office an 
impressive array of books, including biographies and recent nonfic- 
tion bestsellers, that reflected some of the political developments in 
foreign countries whose nationals were sometimes encountered in 
the courtroom. She said she tries to do a fair amount of such read- 
ing, in order to have a better "feel" for the cases that come before 
him. I worried a bit about the evenhandedness of the readings; most 
dealt with the victims of Marxist regimes, and the judge's readings 
seemed to have led to a special reluctance to return anyone to such 
countries. Nevertheless, this extracurricular reading program is 
admirable — far preferable to the attitude encountered in some deci- 
sionmakers who rested content with whatever information the par- 
ties happened to add to the record. But such helpful approaches 
should not be left to the initiative and energy of individual adjudica- 
tors. We should instead devise a system that provides more system- 
atically and honestly for such learning, and also gives assurance that 
adjudicators develop as balanced a picture as possible from their 
readings. 



matters as part of their standard two-week training course, including presiding over a 
simulated deportation case wherein asylum is the chief issue. But this training can 
only be scheduled when there are enough new judges (about eight) to justify the 
session. Therefore some judges may hear numerous cases before attending the 
sessions. In addition, annual conferences of the judges usually offer some program 
on asylum, at times including presentations by refugee advocacy groups and human 
rights organizations. 

24 ' Report of Meeting Between American Immigration Laivyen Association and Executive 
Office for Immigration Review Department of Justice, 65 Interpreter Releases app. Ill at 
743. 749 (1988) (reporting on AILA-EOIR meeting). 

z-'z See In re M— , 7 I. & N. Dec. 222. 225-26 (1956). 



322 DAVID A. MARTIN 

3. Consistency and Quality Control 

An important component of fairness is consistency of outcomes 
among decisionmakers.^** If judged only by asylum decisions in the 
immigration courts, the consistency goal would appear to be well- 
served, primarily through the mechanism of BIA review. Even now 
when the Board ordinarily sits in two panels, it remains a sufficiently 
small and cohesive body that it can reasonably assure similar out- 
comes for similarly situated aliens. Moreover, because both INS and 
the applicant can appeal to the BIA, the Board is in a position to 
police against both false positives and false negatives.^** 

A substantial body of asylum claims, consisting of the applica- 
tions that are granted in the district office, however, escapes this 
checking process. These applications amounted to 39.1 percent of 
cases decided there in FY 1988, and 27.8 percent as a cumulative 
percentage for cases decided over the past five years.^*' Of course, 
anyone unfairly denied asylum in the district office may renew the 
application in immigration court, where consistent results are more 
likely. The problem thus is one of inconsistent positives — a less dis- 
turbing result, perhaps, than inaccurate negatives resulting in return 
to the home country of a truly deserving asylum-seeker. One might 
possibly argue that false positives are not a genuine problem, that 
they reflect merely the system's commitment toward giving asylum 
seekers the benefit of the doubt. 

Such a view should be resisted. False positives, in the long run, 
also harm the system, in two ways. First, the general patterns in asy- 
lum cases are communicated back to the home country.^*^ An exces- 



2*5 In a well-functioning system, consistency should also go far toward serving 
the goal of accuracy. The claim here is more modest, because of doubts about the 
validity of some of the BIA's doctrines, making it too hard for some nationalities to 
win asylum and too easy for others. See, e.g.. In re Fuenies, Interim Dec. No. 3065, 
slip op (BIA Apr. 18, 1988); In re Maldonado-Cruz, Interim Dec. No. 3041, slip op. 
(BIA Jan. 21, 1988), rev'd, 883 F.2d 788 (9th Cir. 1989). Nevertheless, even within 
such a framework, the value of fairness would be served by assuring that Nicaraguans 
in Miami receive the same consideration as Nicaraguans in Nebraska, and that 
Salvadorans in Texas are treated no more harshly than Salvadorans in California. 

^** This BIA capacity, however, is unfortunately undercut to a certain extent by 
APRU's role. Although APRU is scrupulously careful not to intervene in the quasi- 
judicial proceedings before the immigration judges and the Board, it does 
occasionally reopen discussion with INS on the merits of a case after a deportation 
order is administratively final. To the extent that this leads to a later grant despite 
the EOIR denial, it undermines some of the consistency the BIA attempts to obtain. 

^•♦s See supra Table II following note 150. 

246 5^^ supra note 118 (discussing sociological studies of the role of social 
networks in stimulating or facilitating migration). 



REFORMING ASYLUM ADJUDICATION 323 

sive pattern of false positives can thus help stimulate a larger flow of 
marginal asylum applicants. Such a communication process appears 
to have played a role in the surge of Nicaraguan applications in 
Texas in the winter of 1988-89.^*' Second, when the false positives 
are systematically biased in favor of certain groups, as appears to be 
the case at present, they undermine public confidence in the system 
and perhaps increase the chances that courts will be tempted to over- 
turn accurate denials of other nationalities in an attempt to restore 
some rough parity. False positives, when systematically favoring cer- 
tain groups, also violate the underlying premises of the system. 
Those premises require neutral adjudication, committed to provid- 
ing asylum to those, and only those, sufficiendy at risk of persecution 
in the home country.^*® 

The 1980 regulations were crafted with some attention to the 
problem of false positives. Some advocates had urged that the regu- 
lations allow prompt grants of asylum by district directors in merito- 
rious cases, without referral to the Department of State. The Justice 
Department decided against this approach, in part to assure greater 
consistency by checking overhasty grants by examiners.^*® But this 
limited check never worked with great efficiency, because State had 
no way to follow up on cases in which asylum was granted by the 
examiner in spite of a negative advisory letter. In any event, this 
consistency check has virtually evaporated now, when the majority of 
referred cases come back with a sticker that simply indicates that 
State has nothing to add. 



2<7 See Frelick, INS Seeks Tougher Approach on Asylum, Work Authorization. But Faces 
Legal Challenge, Refugee Reports, Jan. 27, 1989, at 1, 2 (reporting that Associate 
Attorney General seeks to rescind earlier Meese policy that was quite generous to 
Nicaraguans, viewing it as "a contributing factor to the current situation" in South 
Texas). 

2*8 See generally Kurzban, supra note 201, at 115. This careful insistence on 
neutrality and consistency in ruling on asylum applications under INA §§ 208 and 
243(h) would not preclude the granting of temporary residence rights to specific 
groups chosen by the political branches, based on a combination of political and 
humanitarian factors, through EVD or special legislation, for example. But decisions 
of that sort, to shelter a wider category of needy individuals, should be clearly seen as 
such — political decisions rather than quasi-entitlements. Such clarity both avoids 
distortion of asylum adjudication and focuses appropriate responsibihty for such 
safe-haven decisions in the pK>litical branches. 

2*9 This theme was voiced fi-equendy in internal government meetings in which 
the author, then a State Department official, participated in 1980. See generally 
Mashaw, The Management Side of Due Process: Some Theoretical and Litigation Notes on the 
Assurance of Accuracy, Fairness, and Timeliness in the Adjudication of Social Welfare Claims, 59 
Cornell L. Rev. 772 (1974) (describing elaborate quality control systems, based on 
sampling and follow-up, used by some social-welfare adjudication systems). 



324 DAVID A. MARTIN 

Finally, it could be that the Asylum Policy and Review Unit was 
meant to bring greater consistency to this body of decisions. APRU's 
method of operation and limited staffing, however, force AFRU to 
place its greatest emphasis on cases in which asylum was denied in 
the district offices rather than when it was granted. In doing so, 
AFRU unnecessarily duplicates the roles played by other units 
involved in the process. Indeed, the Asylum Policy and Review Unit 
is hard to justify under any vision of sound asylum adjudication pro- 
cess. It adds a layer of procedure, and it seems likely only to confuse 
the guidance that INS examiners attempt to follow in adjudicating 
cases, heightening the risks of inconsistency. After all, those examin- 
ers already are mandated to consider the views of the State Depart- 
ment, and they are bound to follow the legal doctrine developed by 
theBIA. 

Why was APRU created? If inadequacies in State Department 
information led to this step, obviously it would be more effective to 
address the specific deficiencies at State — or else to replace that 
Department with another method of providing up-to-date informa- 
tion. If, instead, direct review through the normal channels (involv- 
ing the immigration judges and the BIA) was failing to assure proper 
outcomes, it would have been far better to address those deficiencies 
directly, rather than by throwing another agency at the problem. 

IV. Proposed Reforms 

A, Specialized Adjudicators 

Any cure, short- or long-term, for the ills afflicting our present 
asylum adjudication system should build upon one central change. 
The United States should create a corps of specialized, well-trained 
professional adjudicators^^ to preside at the asylum adjudication 
proceedings and to make the initial determinations in a single, uni- 
fied procedure, replacing the two wasteful rounds of wholly separate, 
de novo consideration now available. The adjudicators should have 
no other function in the immigration system, nor should they rotate 
to this post from other en forcement responsibilities. This change 

250 Similar suggestions have been made for many years. See. e.g., Anker & 
Posner, supra note 31, at 74 ("To insure that applicants receive the sympathetic 
assisunce necessary to a fair resolution of their claims, asylum cases should be 
separated out from routine immigration cases and handled by specially trained 
officials."); Sei-ect Commission, supra note 69, at 173-74. For a sensitive discussion 
of the rules and advanuges of specialized adjudication systems, see S. Lecomsky. 
Specialized Justice (forthcoming Oxford University Press 1990). 



REFORMING ASYLUM ADJUDICATION 325 

would greatly improve the system's accuracy, fairness, and speed, 
whatever other reforms might ultimately be adopted. The change 
would better equip us to welcome refugees promptly, as our tradi- 
tion demands, as well as to deport unqualified applicants expedi- 
tiously. Nearly all other Western countries have built their systems 
around such specialists. It is high time that the United States joined 
their ranks. 

To attract high-quality professionals, the new office should set 
the grade and salary of the adjudicator's position at a level equivalent 
to that enjoyed by immigration judges.^*' Asylum adjudicators make 
decisions every bit as complicated and challenging — and as impor- 
tant to the government and the litigants — ^as other cases that fall 
within present EOIR caseload. This means that costs for the new 
system, at least for the first several years, will run considerably above 
the recent experience under the current system, particularly because 
several dozen adjudicators will be needed.^*^ 

We must develop the capacity, however, to accept short-term 
expenditures in order to avoid larger long-term societal costs, costs 
that are unavoidable as long as we remain vulnerable to influxes of 
marginal asylum applicants. The new system, if effectively imple- 
mented, should finally straighten the many kinks that now prevent or 
^eady delay actual deportation of unsuccessful applicants. Once it 
becomes widely known that the system has that capacity, future 
influxes should decline significandy (barring major outbreaks of per- 
secution in this hemisphere). Such results are, doubdess, years away. 
Expending enough now to do the job right, however, is an indispen- 
sable investment. 

Canada's recent experience reveals the need to consider the 
long-range effects of the expenses. In January 1989, Canada imple- 
mented a promising new system, built around a corps of independ- 
ent, professional, full-time adjudicators supported by its own 
documentation center.^*' At that time, the system was confronted 
with a backlog of 85,000 cases and an intake level that had run to 
45,000 in 1988. To deal with this task, by the summer 1989, the new 
Immigration and Refugee Board had hired not only the sixty-five 



251 Early internal papers that ultimately led to the August 1987 proposed rules 
discussed an option of making the new adjudicators "attorney examiners" with the 
higher rank that classification carries. Unfortunately. INS chose the cheaper option, 
making the asylum adjudicator's position a journeyman-examiner position, at the GS 
11 or 12 grade. 

2*2 See Appendix (for the cost estimates). 

25' See sources cited sufira note 112. 



326 DAVID A. MARTIN 

permanent full-time members authorized by statute for its refugee 
division, but also an additional eighty or more members with two- or 
three-year terms. It also planned to add another fifty or more over 
the next year to help eliminate the backlog. All such officials were 
hired at a high rank, with a salary in the range of $50,000 to $60,000. 
Canada expected to spend over $70 million in 1989 on asylum adju- 
dications.^^ This outlay is considerable. The proposal offered here 
differs from the Canadian model in several important particulars, 
partly in order to develop a less costly system. Significant expense, 
however, cannot be avoided. 

The most important qualifications to be sought in recruiting and 
selecting asylum adjudicators are interest in international affairs and 
demonstrated awareness of and sensitivity to life in other cultures.^** 
Although current adjudicators could, of course, apply for such posi- 
tions, an effort should be made to assure a wide diversity of back- 
grounds among those hired, both to provide the necessary cross- 
cultural sensitivity and to signal that the new system marks a clean 
break from a problematic past.^*^ 

2M Interview with Jerry Robbins, Director General. Operational Policy and 
Planning, Immigration and Refugee Board, in Ottawa, Canada (Dec. 19. 1988). The 
figures are translated into US dollars. 

2*5 Several persons interviewed for the study (most of whom were from 
countries other than the United Sutes) stated that characteristics other than legal 
training may be most important in identifying good asylum adjudicators. They urged 
that serious thought be given to hiring nonattomeys. Canada has taken this 
approach for the majority of the adjudicators (Immigration and Refugee Board 
members) recruited for its new system, even though the position carries a salary that 
would be sufficient to attract attorneys. See sufrra note 254 and accompanying text. In 
an early draft of this study. I thus suggested recruitment among nonattomeys with 
the requisite international experience and sensitivity, without, of course, precluding 
lawyers. This suggestion, however, drew a strongly negative reaction, from 
government officials and refugee advocates alike. Not only will the adjudicators have 
to follow fairly complicated developments in the burgeoning American case law on 
asylum, but also, and perhaps more importantly, they will very often be dealing with 
bwyers for the applicants, in the course of conducting the proceedings and 
constructing the factual record. That record, in turn, is meant to be the basis for 
both the initial decision and further review, which will be entirely under the 
stewardship of lawyers. I have therefore withdra»vn the suggestion. At least for the 
American administrative context, dominated as it is by lawyers and infused with the 
American cultural preference for an adversarial form of justice, asylum adjudicators 
should be trained attorneys. 

256 A. provision in an early Simpson-Mazzoli immigration reform bill would have 
created a separate unit of immigration judges to decide asylum cases after special 
training, but, in an extraordinary measure apparently meant to demonstrate a 
complete break with the past, it provided that no one who had served as an 
immigration judge before the date of enactment could hear asylum cases. See S. 2222, 
97th Cong., 2d Sess. § 124(a)(2), 128 Cong. Rec. 21.671. 21.675 (1982) (proposing 
new Immigration and Nationality Act § 208(a)(2)). The 198S version of the bill 



REFORMING ASYLUM ADJUDICATION 327 

A demonstrable change from the past, fortified by a visible com- 
mitment of added resources to assure professionalism, would also 
serve other useful functions. First, such a change would maximize 
the chances of gaining the support, or at least the acquiescence, of 
refugee advocacy groups. (Indeed, a major effort should be under- 
taken to encourage NGO participation in shaping the final details of 
any such plan.) The August 1987 proposed regulations foundered, 
in part, because they did not sufficiently indicate a genuine depar- 
ture, nor did they reflect any real commitment of new resources. 

Second, a major shift to an impressively professional group of 
adjudicators might also send an important message to the courts. 
The shift would show that the new system was not cobbled together 
solely as a hasty reaction to the recent rise in the numbers of asylum 
seekers, but that the reform attended to the needs of asylum seekers 
truly in danger of persecution at home. The history of dealings 
between agencies and courts in the asylum field suggests that a fresh 
start is well-advised.^^' The object of these reforms is not to launch 
a system that will work only after years of paralysis resulting from 
test cases. The object is to create a system that can work fairly and 
efficiently after only a brief start-up period, and that can actually lead 
to swift grants and denials — the latter leading to prompt deporta- 
tion. This objective cannot be achieved unless the courts are pre- 
pared to defer to the agency in the vast majority of cases.^*® 

Many Western countries have developed adjudication staffs 
whose members specialize by region or even by country of origin. 
Such specialization would be ideal, for it would improve still further 
the development of expert and detailed knowledge to be brought to 



eased this restriction, however, simply preventing such service until the judges 
received special training in international relations and international law. See S. 529, 
98th Cong., 1st Sess. § 124 (a)(1). 129 Cong. Rec. S6970, S6975 (daUy ed. May 18. 
1983). 

2^^ See cases cited supra note 3 (describing the years of litigation which resulted 
when Haitian asylum seekers initiated the first major challenge to the government's 
procedures for asylum adjudication). In the Haitian Refugee Center litigation, the 
element most damaging to the government's case may have been the revealing fact 
that INS tally sheets for reports on the "Haitian program" contained room only for 
the number of denials; it had no line for reporting on grants of asylum. See Haitian 
Refugee Center v. Smith. 676 F.2d 1023. 1031-32 (5th Cir. Unit B 1982). Not 
surprisingly, the courts concluded that the effort was meant to clear dockets without 
attention to the merits of individual cases. See id. at 1040. 

258 See, e.g., Mashaw, supra note 249. at 806-07 (suggesting that an imporunt 
factor determining whether a reviewing court will take an interventionist or a 
deferential stance is the court's "confidence or lack of confidence in the integfrity of 
the underlying administrative process"). 



328 DAVID A. MARTIN 

bear in the cases.^*^ Nevertheless, such a system is probably not 
fully attainable in this country, owing primarily to geography. In 
Switzeriand, for example, the full federal government adjudication 
staff can remain in Berne. Applicants simply come to the capital via 
an inexpensive train ride for their interviews with the appropriate 
country specialist. The United States is too vast for such a system. 
Nevertheless, there may be some chance for a limited specialization 
in various locations where particular nationalities have congregated 
(such as Poles in Chicago or Nicaraguans in Miami). In any event, 
most current asylum seekers come from Central America and the 
Caribbean. Recruitment efforts should therefore focus on persons 
already familiar with the cultures of that region, and all adjudicators 
should receive training and ongoing information on country condi- 
tions there. 

The new system should be expressly based on an understanding 
that these adjudicators will develop expertise about country condi- 
tions over time and may apply their cumulative learning to each case 
they encounter. As developed above, such expertise will help serve 
several important objectives. It will facilitate adequate questioning 
at the hearing to cover all necessary details, help in assessing credi- 
bility, and undergird the ultimate evaluation of the risk the applicant 
would face if returned. Training must emphasize that the adjudica- 
tor's mission is as much to help substantiate meritorious claims as it 
is to issue prompt denials when the claimant is unqualified. 

Other elements, some detailed in succeeding sections, would 
also serve to develop and preserve the needed expertise. Perhaps 
most important would be a well-staffed documentation center, 
independent of the Sute Department. The task of this documenta- 
tion center would be to amass unclassified information on country 
conditions from a wide range of sources, including both the State 
Department and private human rights organizations, in order to 
make this information available in as accessible a form as possible. 
Many other nations have devoted resources to official documentation 
centers of this type. Canada, in particular, has pioneered several 
innovative and useful techniques, including frequendy updated 

259 See, e.g., Aleinikoff, supra note 89, at 254 (advocating a system where 
decisionmakers can improve their ability to judge the credibility of an application by 
becoming "thoroughly familiar with conditions, events, political parties, and social 
groups" in particular countries); Kalin, s^a note 107, at 239 (recommending that 
officials be knowledgeable of the particular cultural background of the asylum- 
seekers whose cases they decide, in order to prevent cross-cultural 
miscommunication and impaired decisionmaking). 



REFORMING ASYLUM ADJUDICATION 329 

country profiles and backg^round information on all significant 
source countries. ^^ In addition to assuring that they remain current 
on developments in source countries, adjudicators should be able to 
use the center, with the help of its staff, to search for information 
about particular legislative facts. For example, if the claimant asserts 
that she was involved in a major demonstration in November 1983, 
in the capital city and that the demonstration was violently sup- 
pressed by the police, an adjudicator could seek confirmation of such 
an event from the center. Or, if the claimant asserts that government 
soldiers forcibly impress young men into the armed forces, the adju- 
dicator could ask the documentation center staff to provide whatever 
information is available from its database on such matters. (Fairness 
constraints on the use of such information are discussed below.) 
The center's resources should also be open, of course, for use by 
asylum seekers and their counsel.^^* 

Because this new system would develop its own capacity for 
obtaining and evaluating a wide range of country condition informa- 
tion, routine referral of cases to the State Department should be 
eliminated.^^^ Individual adjudicators might still refer a particular 
matter when it appears likely that State Department information, not 
otherwise available through the documentation center, would be 
particularly helpful. But solicitation of State Department views 
should be the exception, not the rule. 

There remains the "coast of Bohemia" problem. Indeed, when 
I described early versions of this proposal to some private attorneys, 
they expressed deep concern that expert adjudicators of the sort 
proposed here might become overly dogmatic in their own distinc- 
tive views of conditions in the home country. And with the removal 
of any second-round de novo consideration before the immigration 



260 See Rusu, supra note 112. See generally Rudge & Kjaerum, The Informatum 
Aspects of Refugee Work — Time for a FuU-Scaie Information Strategy, Refugee Abstracts, 
Dec. 1988, at 1; Thoolen. Refugees and Information Technology, Refugees, Oct. 1988, at 
34, 36 (advocating an international refugee documentation network). 

261 Ideally, the new system would follow a practice carefully adhered to in 
Switzerland. The Swiss system specifically provides that 10 percent of each officer's 
work week be set aside for keeping current on the latest information received in the 
center regarding countries for which she has responsibility. 

262 Arrangements should still be made for referral of "urgent action cases," 
such as those involving defectors and diplomats, to the State Department so that it 
may handle the immediate diplomatic consequences. Or perhaps only basic 
biographic data on each applicant could be provided to the State Department, solely 
for the purT>ose of screening by the national security agencies, rather than generating 
a letter containing State Department views on each application. 



330 DAVID A. MARTIN 

judges, advocates would lose at least one important opportunity to 
correct for such bias. 

No system can be designed that escapes this problem altogether. 
But it hardly seems prudent to retain multiple layers of de novo con- 
sideration on the chance that the pictures inside the head of the sec- 
ond adjudicator will cancel out those inside the head of the first. 
Such a system is a recipe for stalemate or confusion, even though 
fewer deportations may actually result while litigation drags on. A 
better approach, though hardly foolproof, is to craft a system can- 
didly aware of the risk of such distortions and dedicated to avoiding 
them. 

TTie best cure for dogmatic stereotyping is steady provision of 
reliable information — constandy forcing the participants to redraw 
the pictures inside their heads to conform more completely to the 
new, more detailed, and more accurate information.^®' This kind of 
corrective is far more likely to be successful under the proposed sys- 
tem than under the present system, if only because the adjudicators 
will be confi'onting such information full time, rather than consider- 
ing asylum cases as a fraction of their workload. Other strategies, 
such as weekly regional updates prepared by the documentation 
center staff, ongoing training procedures, and well-targeted moni- 
toring, can offer further assurances, if pursued with sufHcient deter- 
mination by the agency. 

B. Organizationai Location 

Who exacdy would perform such monitoring and arrange for 
ongoing training? That is, where would this new office be located, 
and what would be its lines of accountability? Elarlier studies have 
sometimes argued for a fully independent asylum adjudication office, 
headed, for example, by a multi-member board appointed by the 
President and removable only for cause. ^^ Independence would 
indeed carry perceptible advantages. Principsdly, it would ensure 
reasonably neutral decision making, insulated from foreign policy 
influences and sheltered from dominance by an enforcement 
perspective. 

But full independence of this sort might not provide adequate 
assurance to those who worry primarily about asylum as a loophole. 

265 See W. LippMANN. suffra note 81. at 402-405. 

2fr» See Aleinikoff. supra note 89, at 234-35 Sc n.l70; see also N.L. Zucker & N.F. 
ZucKER, sufira note 61, at 276-77 (citing numerous studies advocating a refugee 
authority independent of political considerations). 



REFORMING ASYLUM ADJUDICATION 331 

And, in any event, full independence is not likely to be politically 
acceptable. Traditionally, Congress has insisted upon keeping a 
wide range of functions and authorities in the immigration field, 
whether they relate to enforcement or adjudication, under the con- 
trol of the Attorney General. 

It remains possible, however, to achieve most of the objectives 
of independence while retaining general responsibility in the Attor- 
ney General. The obvious location for a new unit of this sort would 
be within the Executive Office of Immigration Review, Created in 
1983, that Office has evolved over the last several years into a major 
subunit of the Justice Department concerned exclusively with adjudi- 
cation and review, and freed of entanglement with enforcement 
functions. Although EOIR reports to the Attorney General, it is suf- 
ficiently removed fi-om foreign policy and enforcement responsibili- 
ties to afford reasonable assurance of neutrality and independence in 
asylum adjudications.^^^ The August 1987 proposed regulations 
might have been more acceptable if they had placed their new corps 
of asylum adjudicators within EOIR, rather than keeping the unit in 
INS. 

The proposal, therefore, could be framed in this way. Statute or 
regulation should create within EOIR a new Asylum Board, headed 
by a chairperson responsible directly to the Attorney General. The 
chair would supervise a staff of asylum adjudicators, hired as 



265 As a practical matter, the Attorney General exercises his authority over 
EOIR decisions only through use of her "referral" power under 8 C.F.R. § 3.1(h) 
(1989) — for example, because it would set a wide-ranging precedent. Referrals are 
rare, and in any event they are publicly known and visible, thus minimizing the risk of 
improper invasion of adjudicative neutrality. See also United States ex reL Accardi v. 
Shaughnessy, 547 U.S. 260 (1954) (limiting other opportunities for control by the 
Attorney General of Board of Immigration Appeals decisions). Similar procedures 
should likewise shield the determinations of the asylum unit. 

Some commentators appear to assume that any adjudications still under the 
responsibility of the Attorney General will be inevitably tainted with an enforcement 
outlook. See, e.g.. Developments in the Law — Immigration Policy and the Rights of Aliens, 96 
Harv. L. Rev. 1286, 1363-66 (1983) (asserting that the "implicit threat of abolition" 
of the Board of Immigration Appeals by the Attorney General "undermines the 
independence of the Board's judgment"). See generally U.S. Comm'n on Civil Rights, 
The Tarnished Golden Door: Civil Rights Issues in Immigration 40-43 (1980) 
(citing several explanations for the emphasis on enforcement activities by the 
Immigration and Naturalization Service). This approach is too mechanical. 
Functional independence and neutrality grow from and are nourished by a far wider 
range of ties and reinforcements; judgments about independence based solely on 
inspection of an organization chart or tenure protection provisions are likely to be 
misleading. See J. Mashaw, Bureaucratic Justice — Managing Soclu. SECURrrv 
DiSABiLmr Claims 41-44 (1983); T. Aleinikoff & D. Martin, supra note 3, at 91, 
451-53. 



332 DAVID A. MARTIN 

described above, who would probably be located in several offices 
distributed around the country as caseloads require. The chair 
would also be a member of any administrative appellate unit dealing 
with asylum (to be discussed below). 

C. Nature of the Proceedings 

Because the asylum adjudicators would lack jurisdiction over 
other immigration law issues, under this proposal, asylum determi- 
nations will obviously be separate from deportation or exclusion 
proceedings. Those who walk in to a district office to apply affirma- 
tively for asylum should receive the necessary form and be given 
enough time to complete it and gather any desired supporting infor- 
mation. Once the form is returned, the case can be referred to an 
adjudicator. Denial by the adjudicator would foreclose future con- 
sideration of the issue in deportation proceedings. If the asylum 
issue is raised only after the alien is already in proceedings, the 
immigration judge should adjourn the hearing, pending a decision 
on the asylum claim by the specialized adjudicator.^^ Alternatively, 
special arrangements could be made, particularly when the party 
concedes deportability and suggests no other relief from deporta- 
tion, for speedy entry of a conditional deportation order— condi- 
tional on the outcome of the asylum adjudication. Careful thought 
should be given to streamlining these procedures so that, if the asy- 
lum claim is not accepted, a fully effective deportation order can take 
effect as soon as possible. 

Other questions about the nature of the proceedings are more 
basic, for they go to the fundamentals of how evidence will be 
presented and tested, both in the interest of the applicant and in the 
interest of the government. Recommended here is a nonadversarial 
model that assigns to the adjudicator the major responsibility for 
developing the record, including the marshaling of both positive and 
negative information, and adds certain measures to assure fairness 
for the applicant and a complete opportunity to present her best 
case. 

The choice of a nonadversarial model may seem surprising in 

266 The August 1987 proposed regulations spelled out detailed arrangements 
for such a procedure. They also provided possible procedural models for 
consideration of asylum claims that arise only after deportation proceedings have 
ended, and for limited opponunities for reopening denied claims based on changed 
circumstances. See 52 Fed. Reg. 32.552. 32.554, 32.558-59 (to be codified at 8 C.F.R. 
pts. 208.8. 208.18) (proposed Aug. 28. 1987). 



REFORMING ASYLUM ADJUDICATION 333 

light of constitutional due process considerations.^^' Under a 
Mathews v. Eldridge^^^ analysis, it is customary to consider that asylum 
seekers have a most vital private interest at stake. And, although the 
government's interests may also be weighty, particularly in light of 
the pressures created when asylum applications are numerous, it 
remains customary to think of adversarial trial-type proceedings as 
the best guarantees — perhaps indispensable guarantees — ^when indi- 
vidual stakes are high.^®^ 

But a closer look at the Supreme Court's procedural due process 
jurisprudence reveals that the Court does not prescribe adversarial 
procedures as a requirement in all settings where important interests 
are at stake.^'° Fundamentally, due process requires the opportunity 



2^^ See generally Scanlan, Asylum Adjudication: Some Due Process Implications of 
Proposed Immigration Legislation, 44 U. Prrr. L. Rev. 261 (1983) (analyzing a number of 
constitutional difficulties with the changes proposed by the Senate version of the 
Simpson-Mazzoli "Immigration Reform and Control Act of 1981"). 

268 424 U.S. 319 (1976). Etdridge established the Supreme Court's framework 
for resolving procedural due process issues. This framework requires courts to 
consider three £actors: 

First, the private interest that will be affected by the official action; second, 
the risk of an erroneous deprivation of such interest through the 
procedures used, and the probable value, if any, of additional or 
substitute procedural safeguards; and finally, the Government's interest, 
including the function involved and the fiscal and administrative burdens 
that the additional or substitute procedural requirement would entail. 
Id at 335. This analysis has often been criticized as inadequate, primarily for focus- 
ing too much on accuracy and too little on the "dignitary" interests of the individuals 
involved. See, e.g., Mashaw, The Supreme Court's Due Process Calculus for Administrative 
Adjudication in Mathews v. Eldridge: Three Factors in Search of a Theory of Value, 44 U. 
Chi. L. Rev. 28 (1976) (asserting that "the fsuling o{ Eldridge is its focus on questions 
of technique rather than on questions of value"). Those critiques may have less 
weight in the immigration setting, but, in any event, Eldridge remains the governing 
standard. 

269 Sgg Mashaw, supra note 249, at 772, 775 (describing and criticizing this view). 
See generally Martin, Due Process and Membership in the National Community: Political Asylum 
and Beyond, 44 U. Prrr. L. Rev. 165 (1983) (arguing for greater acceptance of 
nonadversarial procedures). 

270 See, e.g., Walters v. National Ass'n of Radiation Survivors, 473 U.S. 305, 
320-334 (1985) (finding that a veteran's difficulty in procuring legal representation 
for Veteran's Administration benefit claim procedures caused by a $10 attorney's fee 
limitation does not violate due process requirements); Parham v. J.R., 442 U.S. 584, 
606-09 (1979) (holding that an adversary proceeding to determine the 
appropriateness of decisions to commit children to state mental hospitals is not 
required where an in^iry has been made by a staff physician); Goss v. Lopez, 419 
U.S. 565, 583 (1975) (stopping short "of construing the Due Process Clause to 
require, countrywide, that hearings in connection with short suspensions [from 
public school] must afford the student the opportunity to secure counsel, to confront 
and cross-examine witnesses supporting the charge, or to call her own witnesses to 
verify her version of the incident"); Richardson v. Perales. 402 U.S. 389. 410 (1971) 



334 DAVID A. MARTIN 

to be heard "at a meaningful time and in a meaningful manner."''^' 
What is meaningful should be decided, not in the abstract, but only 
after careful attention to the specific adjudicative task at hand.^^^ 
Eldridge is not to the contrary. In fact, the Eldridge analysis — ^and par- 
ticularly its middle factor — asks us to move away from rigid reliance 
on the classical trial-type hearing model and to inquire instead into 
what makes the most sense for assuring fairness in the precise adju- 
dication at issue. That middle factor invites us to undertake a careful 
comparative inquiry, weighing the relative merits of the adversarial 
and nonadversarial models m the asylum context. Viewed in this light, 
an adversarial asylum hearing, presided over by a passive judge who 
officially knows nothing about the relevant issues except what 
appears in the record, should be seen as a poor servant of either 
fairness or accuracy. 

First, several of the basic assumptions that underlie our usual 
preference for trial-type procedure do not apply here. That prefer- 
ence derives from the view that rebuttal evidence, cross-examination, 
and confrontation provide "the best way to resolve controversies 



(approving significant role for presiding administrative law judge in questioning 
applicant for social security disability benefits, as against claim that this practice 
unconstitutionally mixed the role of prosecutor and judge and upholding use as 
evidence of written physicians' reports supporting nondisability, notwithstanding the 
reports' hearsay character, the absence of cross*examination, and the directly 
opposing testimony by claimant and her medical witness). 

S7I Armstrong v. Manzo, 380 U.S. 545. 552 (1965). 

*72 5^, e.g., Califano v. Yamasaki. 442 U.S. 682, 696 (1979) (noting that the due 
process clause is not necessarily violated when "some leeway for practical 
administration" of a statute is allowed). Many other writers, including two eminent 
federal judges, have urged that American due process inquiry expand its horizons 
and acknowledge that fairness can often be well served by procedures other than 
trial-type hearings. See, e.g., Frankel, The Search for Truth: An Umptreal View, 123 U. Pa. 
L. Rev. 1031, 1052-55 (1975) (challenging the strictly adversarial nature of the U.S. 
legal system and suggesting more investigation into alternative systems of justice); 
Friendly. "Some Kind of Hearing, " 123 U. Pa. L. Rev. 1267. 1287-91 (1975) (advocating 
an investigatory model for administrative hearings under which an independent 
decision maker would have the responsibility for developing all the pertinent facts); 
Langbein, The Cnmtnal Truil Before the Lauryers, 45 U. Chi. L. Rev. 263, 314-16 (1978) 
(explaining that much of U.S. criminal trial procedure, most notably the active roles 
of prosecution and defense counsel, is of only recent historical origin and peculiar to 
the American legal system); see also Parham v. J.R., 442 U.S. 584, 606-09 & n.l6 
(1979) (approving nonadversarial procedures in the context of parental commitment 
of children to state meyal institutions because "[t]he judicial model for factfinding 
for all constitutionally protected interests, regardless of their nature, can turn 
rational decisionmaking into an unmanageable enterprise"); O'Barr 8c Conley, 
Litigant Satisfaction Versus Legal Adequacy m Small Claims Court Narratives, 19 Law & Soc'v 
Rev. 661. 666-67 (1985) (criticizing the restraints placed on witnesses in formal 
adjudication). 



REFORMING ASYLUM ADJUDICATION 335 

involving disputes over adjudicative facts. "^^^ But cross-examina- 
tion and confrontation are rarely among the tools used by an asylum 
seeker in an asylum proceeding, for a fundamental reason: The gov- 
ernment offers its own witnesses only on rare occasions. In the over- 
whelming majority of asylum cases, the only witness actually 
testifying (as opposed to communicating in some fashion through 
documents) is the apphcant himself, perhaps joined by family mem- 
bers. Therefore, the only cross-examination that takes place, most of 
the time, is that of the trial attorney who endeavors to expose incon- 
sistencies or weaknesses in the applicant's own account. It could 
hardly be thought unfair to the applicant to replace such interroga- 
tion (designedly adverse) with questioning done instead by an exam- 
iner who has been instructed that her role is to develop a full record 
and not to strive zealously for a negative outcome. 

The other information in the record is usually documentary, 
such as newspaper accounts or human rights reports. Very litde of it 
relates specifically to the individual; virtually all of it has to do with 
legislative facts. ^'* Reports concerning legislative facts may, of 
course, be rebutted, and occasionally it will be in the applicant's 
interest to attempt to do so— for example, to challenge something 



27' 3 K. Davis, supra note 93, i 15.3, at 144. Professor Davis amplifies those 
reasons as follows: 

The reason we use trial-type procedure. I think, is that we make the 
practical judgment, on the basis of experience, that taking evidence, 
subject to cross-examination and rebuttal, is the best way to resolve 
controversies involving disputes over adjudicative facts, that is. facts 
pertaining to the parties. The reason we require a determination on the 
record is that we think fair procedure in resolving disputes of adjudicative 
facts calls for giving each party a chance to meet in the appropriate 
fashion the facts that come to the tribunal's attention, and the appropriate 
fashion for meeting disputed adjudicative facts includes rebuttal evidence, 
cross-examination, usually confrontation, and argument (either written or 
oral or both). The key to a fair trial is opportunity to use the appropriate 
weapons (rebutul evidence, cross-examination, and argument) to meet 
adverse materials that come to the tribunal's attention. 

Id. 

274 The only likely exception may be an account of earlier statements the 
individual applicant gave to immigration officials, usually at the time of 
apprehension. For example, the account may say that the applicant told the officers 
she came to find a job. If the applicant disputes the accuracy of that account, it may 
be necessary to call^e recording official to the hearing. But most often, the 
applicant does not challenge the fact of the earlier statements; she challenges their 
significance or seeks to explain them because of her reluctance to touch on risky 
subjects in the presence of uniformed officers. Here, too, the applicant's own 
testimony will be the relevant tool, not confrontation or cross-examination. For 
discussion of "legislative facts" see sufrra notes 93, 99-105 and accompanying text. 



336 DAVID A. MARTIN 

asserted in the State Department's human rights country report. But 
again, the usual adversarial tools are not necessarily of great assist- 
ance here. Rebuttal customarily takes the form of supplying compet- 
ing documentary evidence that tends to controvert the asserted fact. 
The nonadversarial model proposed here makes fiill allowance for 
such submissions, along with argument based thereon. Moreover, 
unlike the present system that makes regular provision only for State 
Department input, the model here places responsibility on the adju- 
dicator to consider not only State Department information, but also 
human rights reports from other reasonably available sources. 

Second, an adversarial model functions well only when each of 
the three key roles (the judge plus the parties' representatives) is 
played by a professional who is well-equipped to deal with the sub- 
ject matter and the techniques at hand. In asylum adjudication, satis- 
fying this prerequisite is not simply a matter of assuring the presence 
of lawyers, for the ordinary generalist lawyer's tools often are insuffi- 
cient to carry out an adequate inquiry, even into the immediate adju- 
dicative facts. Substantial country expertise, supplemented by 
sensitivity to cross-cultural difficulties, is necessary even to perform 
an effective direct examination of one's own client. One manual for 
volunteer attorneys in asylum cases illustrates the need for such 
qualities: It recounts the story of "one lawyer who, upon hearing 
that her client had been chased by armed men in civilian clothing in 
El Salvador asked, 'Well, why didn't you go to the police?' "^^* 

In short, the relevant expertise, detailed knowledge about con- 
ditions in source countries, is simply too scarce. One cannot expect 
three participants in the adversarial proceedings (two lawyers and 
one judge) to have this expertise except in rather unusual circum- 
stances. Adopting a nonadversarial model would allow us to target 
resources on making sure that the one key participant, the adjudica- 
tor, is well equipped— equipped not only to make the final judgment, 
but also to frame questions throughout the hearing that will promote 
accurate understanding of the adjudicative facts at issue. 

The nonadversarial procedures in asylum cases should thus pro- 
ceed roughly as follows. The applicant would have the opportunity, 
as at present, to provide whatever information she wished when fil- 
ing the Form 1-589 (or preferably, a better designed application 
form). She might choose to supply lengthy answers to the form's 

275 COMM. FOR HEALTH RiCHTS IN CeNT. Am. AND THE FATHER MORIARTY CeNT. 

Am. Refugee Program, PoLrncAL Asylum: A Handbook for Legal and Mental 
Health Volunteers 44 (n.d.). 



REFORMING ASYLUM ADJUDICATION 337 

questions, submit supplementary affidavits or accounts, or file gen- 
eral human rights information on country conditions. If the adjudi- 
cator is not already familiar with conditions in the source country, 
she will be responsible for establishing such acquaintance, with the 
aid of the documentation center, in advance of the proceedings. 
Such preparation would, of course, include review of all material 
supplied with the application. 

At the actual hearing, the applicant should first be invited to 
recount the important elements of her case and to add anything she 
wishes. The adjudicator would then pose questions meant to flesh 
out the account as necessary, to test its consistency, and to home in 
on the issues that appear, under the facts of the particular case, to be 
crucial to the ultimate judgment about risks faced in the home coun- 
try. No government counsel would appear. If it developed that fur- 
ther information had to be gathered to enable effective examination, 
the adjudicator could adjourn the proceeding. But such postpone- 
ments should be rare. The proceedings should be recorded verba- 
tim, as occurs now m immigration court.**'** 

If the asylum seeker has a lawyer (for example, through the 
efforts of an NGO), counsel could of course be present to advise and 
reassure the applicant throughout the proceeding. Beyond this, 
counsel's role should supplement that of the adjudicator, by posing 
further questions to expand or clarify and to put on other evidence, 
in those cases where such evidence is available. Most of the time the 
case would focus only on the factual inquiry, but, in those cases 
where substantial legal issues arise, counsel could, of course, offer 
argument on points of law. 

The proposal is designed for reasonably full and certainly fair 
development of the affirmative case, even for inarticulate asylum 
seekers who appear without counsel, or with counsel insufficiendy 
familiar with asylum cases or home country conditions. ^^' TTie pro- 



276 Eventually, however, it may be possible to find more expeditious ways to 
preserve the record for appeal. See supra note 174 (discussing use of videotapes). 

277 The setting would thus bear many similarities to social security disability 
proceedings, where the presiding administrative law judge is under an affirmative 
duty to develop both sides of the case. See Mashaw, supra note 249, at 779-83. Courts 
have found ways to police this requirement, particularly in instances where the 
applicant appears pro se. See. e.g.. Bluvband v. Heckler. 730 F.2d 886, 892, 895 (2d 
Cir. 1984) (finding that the administrative law judge presiding over the disability 
hearing failed to help the claimant ferret out all the relevant facts and adequately 
develop the record). Some 70 percent of the disability claimants are unrepresented. 
See i K. Davis, supra note 93, § 14:17, at 86; see also Mashaw, supra note 249, at 781- 
82. Nevertheless, the Supreme Court has specifically upheld this structure against 



338 DAVID A. MARTIN 

posal is therefore meant to enable speedy but fair decisions in a 
heavily burdened system, without being entirely dependent upon the 
availability of pro bono efforts from the private bar.^'® If reasonably 
available, however, counsel's role should be welcomed, primarily for 
the way in which prehearing consultation can serve to sharpen the 
issues and especially to encourage reticent applicants to tell the 
whole story. The less the case has been developed beforehand by 
counsel, the more time the adjudicator will probably have to devote 
in order to identify the crucial factual elements on which the affirma- 
tive case rests. But clearly no adjudicator will be able to spend the 
thirty hours or more that private attorneys report spending, on occa- 
sion, to develop the full trust necessary to coax out the whole story. 
This deficiency must be acknowledged. But the system simply can- 
not be expected to go that far, on governmental resources, to help 
bring forth facts that are that elusive. Claimants bear the burden of 
coming forth with the evidence. The system cannot be designed for 
the chance (although it is admittedly real) that in a small percentage 
of the cases such delay and coaxing will unearth a meritorious case. 

D. Fairness and the Treatment of Legislative Facts 

The expert knowledge developed by the adjudicators would be 
used primarily to ask detailed and focused questions, to help evsdu- 
ate the answers received, and to make the required predictive judg- 
ment about future risks the applicant would face in the home 
country. Using expertise in this manner should not present signifi- 
cant fairness difficulties; specialized adjudication is customarily 
assigned to administrative bodies specifically to take advantage of 



due process challenge. See, e.g.. Richardson v. Wright. 405 U.S. 208. 209 (1972); 
Richardson v. Perales, 402 U.S. 389, 410 (1971). 

2^^ As a further measure of reassurance to unrepresented applicants, especially 
those from backgrounds that might make official proceedings intimidating, it might 
be possible to emulate a feature of the Swiss system. The Swiss government pays a 
small stipend to volunteers, recruited by an umbrella refugee assisunce organization, 
who have a right, by statute, to attend each asylum adjudication interview. Ste Loi sur 
I'asile du 5 octobre 1979 § 3, Recueil officiel des lois et ordonnances de la 
Confederation Suisse [RO] 1980, at 1717, 1718, amended Modification du 16 
decembre 1983, RO 1985, at \, further amended Modification du 20 juin 1986, RO 
1987, at 1674 (taking effect Jan. 1, 1988) (codified at Recueil systematique du droit 
federal [RS] 142.31). They are there primarily as observers, and they clearly do not 
see their role as lawyer-substitutes for the applicant; they do not meet with the 
applicants beforehand. But their presence can serve as an additional guarantee of 
fairness, and they also are generally permitted to pose questions at the end of the 
procedure to clear up any confusion or ambiguities. 



REFORMING ASYLUM ADJUDICATION 339 

such cumulative learning and specialization.^^® But if some facts 
developed by the expert adjudicators become central to particular- 
ized determinations crucial to the ultimate ruling, fairness may 
require additional steps before relying on this outside information. 
An example will help to clarify this point. 

Suppose that the applicant claims she will be persecuted because 
she was a local organizer with the XYZ political party, a radical splin- 
ter group operative in a certain province of the home country. She 
offers evidence of a government crackdown on the organization, and 
indeed the country profile from the documentation center likewise 
reports the crackdown. But after examination, the adjudicator is pre- 
pared to rule as follows: 

I find the asylum seeker not to be credible in her claim of 
involvement with the XYZ party. I reached this conclusion primar- 
ily on the basis of certain questions I posed to her. I asked her who 
A was and I asked her who B was, and she did not know. A and B 
are key leaders of the XYZ party in that region (citing the sources 
of this information). Anyone even minimally active with XYZ 
would have known that. Therefore, her testimony regarding 
involvement with that group is not worthy of belief. Because her 
claim rested solely on that ground, her application for asylum %viU 
be denied. 

In Professor Davis's conceptual scheme, this information about 
A and B is a legislative fact because it is not a £ict concerning the 
immediate party. Hence, it need not necessarily be placed on record 
by means of live testimony subject to cross-examination; official 
notice is appropriate.^®® But because the information is being used 

279 See, e.g., E. Gellhom, Rules of Evidence and Official Notice in Formal Administratixte 
Hearings, 1971 Duke L. J. 1, 43 ("In reaching a conclusion, the examiner or agency 
may rely on its special skills . . . just as a judge may freely use her legal skills in 
reading statutes and applying decided cases in the preparation of her opinion."); W. 
Gellhom. Officuil Notice in Administrative Adjudication, 20 Tex. L. Rev. 131, 136 (1941) 
("The conventional process of proof presupposes, in the main, that each case is a 
separate entity, which the trier of fact approaches with a more or less blank mind. 
The hypothetical foundation for that conventional process is absent when the trier of 
fact is an experienced govemmenul agency."). 

Supreme Court precedents also support this notion. See, e.g., FCC v. National 
Citizens Comm. for Broadcasting, 436 U.S. 775, 813-14 (1978) (holding that because 
the determinations were "primarily of a judgmental or predictive nature," the agency 
could apply its expert knowledge; "complete factual support in the record [for the 
agency's conclusions] is not possible or required"); NLRB v. Seven-Up Bottling Co., 
344 U.S. 344, 348-49 (1953) (a board may use its "cumulative experience," which 
"begets understanding and insight by which judgments not objectively demonstrable 
are validated or qualified or invalidated"). 

280 See 3 K. Davis, supra note 93, § 15.10, at 184. Some cases uke a narrower 



340 DAVID A. MARTIN 

here as the crucial basis for a credibility judgment, fairness may 
demand specific notification to the individual, with an opportunity to 
rebut. The Administrative Procedure Act makes provision for such 
situations in adjudications covered by its terms. It provides: "When 
an agency decision rests on official notice of a material fact not appear- 
ing in evidence in the record, a party is entided, on timely request, to 
an opportunity to show the contrary."^®' 

A variant of a procedure now used in district offices, when the 
examiner is prepared to deny an asylum claim based on information 
received from the State Department, might be employed here to 
affijrd the party an opportunity to rebut. Under the reformed sys- 
tem, the adjudicator could simply issue a "notice of intent to deny," 
giving the reasons as described above and citing the source for the 
information about A and B. The asylum seeker would then have a 
specified period of time (district offices now allow fifteen days) to 
rebut the information, either by showing that the source was mis- 
taken and that A and B were not involved with the XYZ party, or by 
providing other reasons why the asylum seeker could not be 
expected to know them. This procedure should be sufficient to sat- 
isfy due process requirements.^®^ 

It should be emphasized, however, that in most asylum adjudica- 

view of ofHcial notice, apparently limiting it to the much smaller range of £acts that 
may be judicially noticed — i.e., facts that are common knowledge or cannot 
reasonably be disputed. See, e.g., Sosna v. Celebrezze, 234 F. Supp. 289. 290 (E.D. 
Pa. 1964) (holding that court's use of medical textbooks to bolster reports of 
examining physicians without notice to claimant was contrary to the Administrative 
Procedure Act); Glendening v. Ribicoff, 213 F. Supp. 301. 302 (W.D. Mo. 1962) 
(holding that the "consideration of . . . extra-record medical information was 
erroneous as a matter of law"). But the better authority is to the contrary, 
acknowledging that a wider scope for official notice is the concomitant of agency 
expertise. See, e.g., McLeod v. INS, 802 F.2d 89. 92-93 (3d Cir. 1986) (approving use 
of official notice in asylum cases and emphasizing that it is a broader concept than 
judicial notice). Some cases taking a restrictive approach to official notice base their 
concern on the fact that such a practice may effectively shift the burden of proof from 
the agency to the individual. See Dayco Corp. v. FTC. 362 F.2d 180. 186 (6th Cir. 
1966); cf. E. Gellhom, nipra note 279, at 45. But in asylum cases, the individual 
clearly bears the burden of proof in any event. Cf. Zamora v. United Sutes, 534 F.2d 
1055, 1062 (2d Cir. 1975) (Friendly, J. dictum) (approving use of State Department 
information in asylum cases, without making drafter of Sute letter available for cross- 
examination, so long as letter speaks only to legislative facts). 

281 5 U.S.C. § 556(e) (1988) (emphasis added). See generaUy 3 K. Davis, supra 
note 93, § 15.10, at 184; E. Gellhom, supra note 279, at 42-49. 

282 It might be possible to justify the issuance of the hypothetical ruling in the 
text even without advance notice of intent to deny, so long as an administrative 
appeal system is available on terms that would permit the alien to file rebuttal 
information in that forum. In the analogous situation under the APA, § 556(e) 
requires only that the opportunity to rebut be made available "on timely request." 



REFORMING ASYLUM ADJUDICATION 341 

dons this procedure will not be necessary. Most decisions will not 
"rest on" official notice of specific legislative facts of this character. 
Instead, the adjudicator will simply employ her general knowledge in 
making the ultimate predictive judgment about the risks the appli- 
cant would face on return. 

E. Administrative Review 

The objective of expeditious proceedings demands that the sys- 
tem achieve final deportation orders quickly, primarily to serve as a 
deterrent against others in the home country with marginal cases 
who may be thinking of coming to the United States to file for asy- 
lum. Obviously, any provision for administrative or judicial review 
will undercut that objective to some extent. Yet to leave the decision 
in the hands of one official, without even a cursory review on the 
administrative record compiled at the initial stage, would be intolera- 
ble when so much is potentially at stake for the individual. Some sort 
of review is indispensable. 

1 . Administrative Review or Not? 

Because of the habeas corpus clause in the Constitution,^®' judi- 
cial review in some form appears inescapable. (Appropriately chan- 
neled judicial review is also highly desirable as an outside check on 
the administrative agency.) It is therefore tempting to consider elim- 
inating administrative appellate review altogether, in the interest of 
speedy finality. After all, if judicial review must be available, then 
any administrative review simply adds a third layer of consideration. 

Canada yielded to this temptation. The 1988 legislation estab- 
lishing its new asylum adjudication system eliminates any centralized 
administrative review by a body equivalent to our BIA. If the United 
States should choose to follow that model (it is not the one favored 
here), it should at least take the other steps Canada pursued to mini- 
mize the risk of inconsistency and error despite the absence of 
administrative review. In Canada, proceedings on the merits of an 
asylum claim are invariably heard by a panel of two members of the 
Immigration and Refugee Board. The asylum seeker needs to per- 
suade only one of them that the case is meritorious in order to 
achieve a favorable result.^®^ 



283 See U.S. Const, art. I. § 9. d. 2. 

2M Sa [Canadian] Immigration Act $§46.02. 69.1(10). ch. 35. §§48.02, 
71.1(10). 1988 Can. Stat. 903, 919. 938; see also sources cited sufna note 112. A 



342 DAVID A. MARTIN 

Nevertheless, several people interviewed in Canada acknowl- 
edged that the new system remains vulnerable to the vice of inconsis- 
tency. For example, if it comes to the attention of the chairman of 
the Immigration and Refugee Board that a panel in Vancouver is 
g^nting asylum readily to members of a certain dissident group, but 
panels in Montreal are consistently denying asylum in such circum- 
stances, the chairman has no direct measure available for achieving 
unified Board policy on the issue. Informal controls, primarily 
through the use of legal opinions or other advice from the Board 
general counsel's office, will probably ameliorate the inconsistency 
problem, but such measures are only advisory. Theoretically, consis- 
tency could be established in Canada through judicial review. But 
judicial review under these circumstances is heard in different courts 
at various levels and in various locations, rather than before a single 
tribunal. In this situation, consistency via the judicial route may take 
a long time to achieve. 

In addition, centralized administrative review is desirable 
because of the difficult nature of the decisions that asylum adjudica- 
tors must make. Tliis difficulty is best illustrated by use of a hypo- 
thetical. Assume that human rights reports reveal a gradually 
increasing pattern of government suppression of labor union activ- 
ists in a Central American country. Tlie first reports mention iso- 
lated arrests of certain leaders. Subsequent reports indicate that 
some of these detained leaders have been tortured. A few weeks 
later a wider circle of prominent union activists are arrested, 
although many still remain at large. At some point, the govern- 
ment's pattern of persecution crosses an important threshold, to the 
point where all union activists found in this country should be recog- 
nized as refugees based solely on their status as union activists. 
Determining when that line is crossed, however, is a difficult judg- 
ment call. In the midst of this evolving pattern, it would not be sur- 
prising for adjudicators initially to reach differing results. 
Consistency would be best served if a centralized forum exists for 
making a definitive and binding decision as to when the line is 
crossed—or at least for assuring that union activist asylum applicants 
in Miami are treated the same as their counterparts in California. 

2. The Recommended Framework 

A reformed U.S. asylum adjudication system therefore should 

helpful codified and annouted version of the Canadian Act is The Annotated 
Immigration Act of Canada (F. Marocco 8e H. Goslett eds. 1989). 



REFORMING ASYLUM ADJUDICATION 343 

retain an administrative appellate body, both to make such difficult 
judgment calls and to monitor for consistent implementation of the 
standards throughout the country. Its basic role would be to con- 
sider appeals from denials by the initial adjudicators; consideration 
would be based on the administrative record and briefs filed by the 
asylum seeker as appellant. The appellant should be allowed a lim- 
ited time, perhaps ten days, for the filing of an appeal. Likewise, the 
time allowed for briefing should be subject to fairly strict limitations. 
Even if the initial stage before the adjudicator is not adversarial, it 
may be worthwhile to treat appeals in a more adversarial manner, 
using INS appellate attorneys (as under the current system) to repre- 
sent the government's interest when the matter reaches the adminis- 
trative appellate body. 

For reasons sketched earlier, it might also be advisable for the 
appellate authority to perform some monitoring role with respect to 
grants of asylum. Although inaccurate grants provoke less concern 
than erroneous denials, a broad pattern of undeserved g^nts serves 
to undercut the public's confidence in the system's fairness. To 
guard against this phenomenon, the staff of the appellate body might 
regularly receive and review decisions in all asylum cases, appealed 
or not, to watch for aberrant patterns. (The staff at this level, being 
centralized in one location, could probably specialize by region or 
country.) In limited circumstances, the appellate body could then 
use the device of certification^®* to bring an unappealed case, posi- 
tive or negative, before it for further review. This sort of monitoring 
would provide a useful quality-control mechanism. 

Thus, the appellate caseload would consist primarily of appeals 
initiated by denied claimants, supplemented by a handful of other 
cases brought before the body on its own initiative. Given adequate 
staffing, and assuming a solution to the transcript problem, this pro- 
cess of review — stricdy on the record created below — should add 
only a few months to the overall delay, and then only in cases 
accepted for full appellate consideration. 

There remains the question of the composition of the appellate 
body. Clearly the current Board of Immigration Appeals could per- 
form this function; approximately half of the BIA's time is already 
devoted to asylum cases. Although the question is a close one, in the 
end, I recommend against assigning these functions to the BIA. 
Fairly or not, some NGOs identify the BIA as a significant source of 

285 See 8 C.F.R. §§ S.l(c), 3.7 (1989). A similar procedural mechanism provides 
for "referral" of cases to the Auomey General. See id. i 3. 1(h). 



344 DAVID A. MARTIN 

the biased results that they believe the system has achieved over the 
past several years. Creation of a new Asylum Board would help sig- 
nal the reality of a fresh start, making the more restrictive elements 
of the new scheme more acceptable. Moreover, asylum is likely to 
generate a substantial portion of contested cases under the immigra- 
tion laws for the foreseeable future, thus justifying the creation of a 
new and permanent unit. 

An Asylum Board, as a separate administrative appellate body 
focusing solely on asylum cases, also will have a better opportunity 
to develop the necessary expertise in the function, including detailed 
acquaintance with home country conditions.*®® Members of the 
board should be attorneys, because difhcult legal questions under 
the asylum provisions of the immigration laws will have to be setded 
in this forum.*®^ The Asylum Board should be located org^ization- 
ally within EOIR, and as indicated earlier, the chairperson of the 
Asylum Board should have general oversight and administrative 
responsibility over the corps of adjudicators.'^*® 

3. Country Guidelines 

Some earlier reform proposals have suggested country guide- 
lines or profiles as a device that would help streamline the process 
and simplify the adjudicative task.*®^ For example, by 1981, it 

286 Xn the press release announcing the creation of APRU, the Justice 
Department emphasized that asylum decisions are "distinct from the normal 
operation and administration of the immigration laws." Attorney General Announces 
New Asylum Policy Unit, 64 Interpreter Releases 472-73 (1987). A new Asylum 
Board would thus allow appropriate specialization on the pan of the BIA. 

287 Recent cases have presented the BIA, for example, with difficult legal 
questions concerning the appropriate standards for asylum claims by conscientious 
objectors, see In re A.G.. Interim Dec. No. 3040, slip op. (BIA Dec. 28. 1987). or by 
participants in a coup plot. See Dwomoh v. Sava. 696 F. Supp. 970 (S.D.N.Y. 1988). 

288 xhe Asylum Board would also provide a logical centralized forum for 
receiving the views of the UNHCR (the Board being the rough equivalent of the State 
Department under the ad hoc arrangements worked out in the 1970s for UNCHR file 
review in Haitian cases, $cc si^a note 191 and accompanying text). UNHCR officials 
suted to me that their preferred point of access is at the administrative appellate 
stage. Interview with Richard Stainsby. UNHCR Senior Legal Adviser, in 
Washington, D.C. (Oct. 1988). Several other countries have made arrangements for 
routine file review by UNHCR officials in this manner. Many NGOs place a high 
priority on a well-targeted UNCHR role, and its expertise could be of genuine 
assistance to the decisionmakers. 

289 See, e.g., Sei-ECT Commission, supra note 69, at 169-171 (recommending that 
"group profiles should be develo|>ed and used by processing personnel and area 
experts to determine the legitimacy of individual claims" and expedite the process): 
Verkuil, supra note 203, at 1 172 (suggesting that a list of countries with clear good or 
bad human rights records be prepared and updated); Scanlan, supra note 2, at 637-38 



REFORMING ASYLUM ADJUDICATION 345 

became clear that the regime in Iran had begun systematic persecu- 
tion, often including summary execution, of adherents of the Baha'i 
faith. The State Department issued a policy statement announcing 
that Baha'is who escaped Iran should be considered refugees ipso 
facto.^^ A similar firm guideline might have been possible in 1978, 
declaring that anyone who escaped the indiscriminately murderous 
policies of Pol Pot's Cambodian government should be considered 
to have a well-founded fear of persecution on return. With such a 
guideline in place, the adjudicators would be freed of responsibility 
for the ultimate predictive judgment about whether the threat level 
crossed the threshold leading to recognition of refugee status. 
Instead, adjudicators would be able to focus on a narrower.and more 
easily ascertainable issue: whether the individual claimant was truly 
an Iranian Baha'i or a Cambodian (and possibly whether she fell 
within one of the exclusion or cessation clauses of the refugee 
definition). 

With the removal of the State Department from any major role 
in refijgee matters, responsibility for discerning such patterns and 
issuing appropriate guidelines to asylum adjudicators — if guidelines 
are to be used at all — would appropriately devolve on the Asylum 
Board. The Board would remain primarily an adjudicative body, but 
the guidelines could be viewed as a natural outgrowth of the regular 
monitoring of country conditions that the Board should perform 
anyway in discharging its adjudicative responsibilities. 

A large dose of realism, however, should curb extensive expec- 
tations about the likely utility of country guidelines. First, in view of 
the current caseload, which comes predominandy from Central 
America and the Carribean, appropriate occasions for their issuance 
are likely to be exceedingly rare. Clear patterns like those occurring 
in the Baha'i or Cambodia examples are unlikely to manifest them- 
selves very often.^®' Guidelines are usefril only when they can be 

(suggesting similar country profiles). The comparison is often made to overseas 
refugee programs, where country guidelines (more in the nature of group 
presumptions of refugee status) are sometimes employed. But such an approach is 
not workable in asylum. Rougher judgments on refugee status are tolerable in 
overseas processing because other screening tools provide an enforceable cap on the 
number who will actually be admitted to the country, however many are initially 
adjudged to meet the reftigee definition. See supra note 38. 

290 See StaU Dept. Reaffirms Policy on Asylum for Iranian Jews. 62 Interpreter 
Reijases 1000 (1985) (describing earher policy announcements on Baha'is, as well 
as on Christians and Jews from Iran). 

291 Some £uro(>ean officials reported that clear patterns appear in a larger 
proportion of their caseload than occurs in the United States. They mentioned 
situations like that in Tiukey, where opposition groups are highly organized and 



346 DAVID A. MARTIN 

based on particularized characteristics that sharply distinguish a cer- 
tain group from the rest of the population. Most persecution in 
countries significantly represented in the current asylum caseload 
does not follow such crisp patterns. If guidelines can only say that 
"prominent" union activists or "visible" governmental opponents 
are likely to be persecuted, the subsequent adjudicative process will 
have to cover almost all the same ground it would cover in the 
absence of guidelines. The adjudicator would still have to pursue in 
detail the applicant's own personal history as a means of judging 
prominence or visibility based on her past activities and any threats 
made against her or her family or friends. Guidelines that must use 
such vague terms are probably worse than no guidelines at all, for 
they would impart an aura of misleading clarity, when the circum- 
stances still require a highly individualized, contextual judgment. 

Second, most guideline proposals envision the use of only 
affirmative guidelines — g^delines that lead to a grant of asylum if 
the individual matches the profile. Negative guidelines verge on 
denying individuals the right to demonstrate that their own personal 
threats are so great that they deserve recognition as 1951 Conven- 
tion refugees regardless of the general state of human rights obser- 
vance in the home country .^^ Although an approach relying only on 
affirmative guidelines is thus understandable, it obviously undercuts 
the utility of guideUnes in streamlining adjudication. Moreover, the 
risk would persist that the absence of an affirmative guideline could 
be taken as an implicidy negative factor by an adjudicator. 

There is a third limitation. Most proposals for the use of guide- 
lines or profiles assume that they would be made public.^' But in 
nearly every country visited, asylum officials expressed great skepti- 
cism about the idea of published guidelines. One Swiss official com- 
mented: "The next week half the applications would match the 
guidelines." Published country guidelines might wind up simplifying 
the ultimate predictive judgment about danger levels at the cost of 
encouraging more sophisticated fi-aud, thus complicating adjudica- 
tion over whether the applicant truly belongs to a class favored by 
the guidelines. Other profiles used for a variety of law enforcement 



government response apparently correlates closely to the precise cell or splinter 
group to which the asylum seeker belongs. 

292 See N.L. Zucker & N.F. Zucker, supra note 61, at 272-73. 

293 See Verkuil, supra note 203. at 1 172. Verkuil goes further and suggests that 
guidelines be adopted through a notice-and-comment rulemaking proceeding. Such 
a process, however, would appear to be too cumbersome to keep up with necessary 
changes as country conditions evolve. 



REFORMING ASYLUM ADJUDICATION 347 

and administrative purposes usually remain a closely guarded 
secret.^®'* But if confidentiality is maintained (except to the extent 
that the underlying information is manifest in written decisions 
explaining individual g^nts and denials), the guidelines could be 
subject to the charge that they amount to a kind of secret and unac- 
countable decisionmaking.^®^ 

On balance, country guidelines probably would cause more 
problems than they would solve. They are by no means essential to a 
well-functioning system, and a reformed American structure proba- 
bly should be designed without provision for them. 

F. Judicial Review 

Under current statutes, asylum determinations are fiilly review- 
able in court, usually in connection with review of a deportation or 
exclusion order under INA § 106 (in the court of appeals for depor- 
tation, in the district court for exclusion). Courts apply either a 
"substantial evidence" or "abuse of discretion" test, depending on 
the precise issue. 

Given overloaded court dockets, these avenues for review create 
significant potential for delay. If most denied applicants were to 
petition for judicial review after exhausting administrative remedies, 
delays would mushroom, negating any effective deterrent message 
that might derive from prompt returns. Although this appears an 
unlikely prospect at present, complete assurance against debilitating 
backlogs might someday require limitation or careful channeling of 
judicizd review, which could be achieved only by statutory amend- 
ment.^®^ But any trimming will be highly controversial, both 

2»4 See]. MoNAHAN & L. Walker. Social Science and Law 207-27 (1985). 

295 This objection was voiced vigorously by the American Bar Association's 
Coordinating Committee on Immigration Law in its comments on a prehminary 
version of this study. Letter from Charles C. Foster to David A. Martin (Apr. 13, 
1989). 

29* Some proposals have been offered that would eliminate judicial review of 
asylum decisions as part of a package of reforms g^rafting several additional 
safeguards onto the administrative process. See, e.g., Scanlan, Isstu Summaries Submitted 
to the Select Commission on Immigration and Refugee Policies by the Center for Civil and Human 
Rights, in Select Commission on Immigration and Refugee Poucy, U.S. 
Immigration Poucy and the National Interest, app. C to the Staff Report (Papers 
on Refugees) at 43, 67 (1981). 

Short of emergency circumstances, attempts to eliminate judicial review are 
inadvisable for two reasons. First, carefully framed, such review plainly can serve a 
most useful checking function, assuring fulfillment of the protective purposes of our 
asylum laws. See generally Legomsky. Political Asylum and the Theory of Judicial Review, 73 
Minn. L. Rev. 1205, 1209-11 (1989). Second, complete denial of review may not be 



348 DAVID A. MARTIN 

because courts have performed a genuinely valuable service by cor- 
recting significant bureaucratic error or abuse in asylum processing, 
and because Americans generally hold the courts in high regard as 
guarantors of rights. Such changes therefore should be considered a 
last resort, to be employed only if the effectiveness of the administra- 
tive changes proposed above is badly undercut by the questionable 
use of judicial review for purposes of delay — abuse that is now rare. 
Thus, the discussion that follows should be taken only as a prelimi- 
nary sketch of possible changes to judicial review — changes which, 
one hopes, will not be necessary. 

Two primary aims, necessarily in tension, are generally accepted 
for judicial review in the asylum scheme. Judicial review should (1) 
play a limited but effective role in checking bureaucratic mistake or 
abuse, and (2) avoid imposing undue delay. With respect to the first 
goal, the court's checking function is necessarily limited; almost no 
one believes feasible a system in which the courts make de novo 
determinations of asylum. With respect to the second, obviously no 
one favors undue delay. Although views may differ on what delay is 
excessive, the discussion in earlier sections points out why expedi- 
tiousness is unusually important in the asylum setting. Delays tolera- 
ble in other administrative settings may become unacceptable here. 

Should changes become necessary, close attention to these two 
aims suggests a reformed judicial review scheme that might maxi- 
mize each. Deterrence of unworthy asylum seekers requires speed, 
but does not require the swift return of everybody who files an 1-589. 
It requires swift return only of those whose cases are at best thin or 
marginal. In all likelihood, such cases constitute a substantial major- 
ity of the current caseload and of reasonably foreseeable caseloads in 

constitutional under article I, § 9 of the Constitution, which provides: "The Privilege 
of the Writ of Habeas Corpus shall not be suspended, imless when in Cases of 
Rebellion or Invasion the public Safety may require it." U.S. Const, art. I, § 9, d. 2. 
See generally Developments in the Law — Federal Habeas Corpus, 83 Harv. L. Rev. 1 038, 
1263-74 (1970). Aliens being removed from the country must necessarily be taken 
into custody, and it would not be difficult in many cases to allege colorable 
constitutional violations — the foundation for issuance of the Great Writ. In any 
event, absolute preclusion statutes tend to spring leaks. See, e.g., Johnson v. Robison, 
415 U.S. 361, 373 (1974) (holding that preclusion-of-review clause in veteran's 
benefits legislation does not bar judicial consideration of constitutional challenges to 
the legislation). A more productive course is to concentrate energies on channeling 
review into forms that will maximize effective judicial checks with minimum 
disruption. When substitute mechanisms are available for review in some form by 
Article III judges, restrictions on the availability of habeas corpus have been held 
valid. See. e.g. Swain v. Pressley, 430 U.S. 372, 381 (1977) (the substitution of a new 
collateral remedy that is neither inadequate nor ineffective does not suspend the 
privilege of the writ of habeas corpus). 



REFORMING ASYLUM ADJUDICATION 349 

time of major influx.^^' Obviously, those with clearly meritorious 
claims must be permitted to stay. But presumably these are the cases 
the new corps of specialists will readily grant, thus obviating judicial 
review. 

This leaves a third category: difficult cases on the boundary, 
understandably requiring more thorough deliberation, research, and 
possible reconsideration. Provided such cases continue to constitute 
a fairly small percentage of the caseload, this category of cases could 
remain pending in the overall administrative-judicial system longer 
without much damage to the deterrent message.^®® If this guess 
about proportions is roughly correct, the system could grant some 
form of access to Article III courts to all asylum seekers, provided 
that the mechanisms permit speedy termination of review unless a 
truly substantial question is raised.^®^ 

Canada's new system provides a potentially useful procedural 
model. Its legislation disallows judicial review of denied refugee 
claims unless the applicant first obtains "leave to appeal" from a 
specified court.'®** This device is not familiar to U.S. lawyers. Our 
system ordinarily allows review liberally without prior screening, 
although meridess appeals may be disposed of summarily. Our 
nearest analogue may be the certiorari process in the Supreme 
Court, which is plainly a screening mechanism that we reserve for the 
highest levels of appellate consideration.*®' Applying such a device 
at the very threshold of judicial review is unlikely to win easy accept- 



*9' This guess would be changed if a Central American Hider or a Caribbean 
Pol Pot came to power. But we would be in a much stronger position to rally political 
support for massive acceptance of refugees in such desperate circumstances if the 
system has won earUer public backing through dear demonstration of the capacity to 
refuse asylum when people do not qualify. 

298 This threefold division of cases (unqualified, difficult borderline, 
meritorious) is in principle a clear one. In practice, of course, actual location of the 
boundaries will be much tougher. If administrators differ greatly from the courts on 
where those boimdaries lie, the latter are bound to intervene more, creating delays in 
more cases and undermining the hoped-for humane and targeted deterrent. 

299 For other proposals to curtail judicial review while maintaining needed 
safeguards, see Aleinikoff, supra note 89, at 236-38. 

300 [Canadian] Immigration Act §§ 82.1-84.1, ch. 35. §§ 83.1-85.1. 1988 Can. 
Sut. 903, 946-49. 

SOI Certificates of probable cause, needed under 28 U.S.C. § 2253 (1988) and 
Fed. R. App. P. 22(b) to appeal a district court's denial of habeas corpus to a state 
prisoner, bear some similarities to this scheme, but there are important differences. 
First, the threshold for issuance is lower than what is suggested here. See Barefoot v. 
Estelle, 463 U.S. 880, 885 (1983). Second, denial of the certificate is itself open for 
reconsideration by a judge of the appellate court, whereas Canada has precluded 
further review of denial of leave to appeal. 



350 DAVID A. MARTIN 

ance. Proponents would carry a heavy burden of persuading the rel- 
evant audiences (including the Congress) that a unique device of this 
type is needed because of the special requirements of an asylum sys- 
tem being challenged by steeply rising numbers of applicants. 

Under the 1988 revisions to Canadian law, asylum applicants 
seeking court review on the merits (including review of assertions 
that natural justice has been violated — the equivalent of our constitu- 
tional due process claims) must file for leave within fifteen days of 
the administrative decision. The court will ordinarily make a deter- 
mination on the application without a personal appearance. If leave 
is granted, the matter is scheduled for full hearing in the ordinary 
course. But if the judge is not persuaded that the case is worth con- 
sidering, the matter goes no further; there is no appellate review of 
denial of leave. What makes a case worth hearing? Unfortunately, 
the Canadian legislation does not specify clearly, leaving it to the 
courts to develop precise standards. '°^ Because the new Canadian 
scheme is still relatively new, it remains unclear just what operational 
tests will come to govern. 

If this "leave to appeal" approach were to be adopted in the 
United States, the statute should clearly state the governing stan- 
dard. The exact formulation requires further attention. But the 
basic idea, if both of the above stated goals are to be served simulta- 
neously, would be to preclude full-fledged court review — with com- 
plete briefing and argument — ^unless there is a substantial likelihood 
of reversal of the administrative action.'*" This is essential. For 
such a change to effect the desired results, Congress would have to 
signal clearly that it expects substantial deference on the part of the 
courts to administrative decisions, and hence expects leave to be 



502 See [Canadian] Immigration Act § 84.2. ch. 35. § 85.2. 1988 Can. Stat. 903. 
949. 

50' This standard comes close to the test applied when a single Justice of the 
Supreme Court considers an application for a stay pending the full Court's ruling on 
the petition for certiorari. See, e.g., John Doe Agency v. John Doe Corp.. 109 S.Ct. 
852. 853-54 (Marshall. Circuit Justice, 1989) (gp^nting stay, in part because there is a 
"fair prospect" that the full Coun would find the decision below to be erroneous). 
Although this formulation makes general schematic sense, it has hardly been framed 
in language suitable for a statute. I am not quite sure what precise formulation 
should be used to get the job done. What is meant to be communicated is more a 
mood or a posture for the courts, rather than a precise schema. The standard should 
signal that most often the job of adjudication belongs to the agency; the courts 
should not intrude too deeply into precise development of substantive standards or 
their implementation in the particular case. Court review is to be used as an outside 
check, an occasional chastener and reminder that accounubility also runs to persons 
outside the bureaucracy. 



REFORMING ASYLUM ADJUDICATION 35 1 

granted only in a small fraction of overall cases. (Moreover, court 
denials of leave to appeal should not themselves be appealable.) If 
the proportions do not work out as sketched above, however, and if 
most cases wind up being heard on the merits in the courts, then the 
"leave to appeal" arrangements, ironically, would actually serve to 
compound delays, by adding an additional round of paperwork. Full 
success depends on both an attitude of restraint by the courts and a 
dedication to high-quality professional adjudication by the agency, to 
reinforce the idea that judicial deference is fully merited. 

If the scheme works as envisioned, all denied claimants would 
have access to an Article III judge; no bureaucratic decision could 
block that access. This fact is vital, for it preserves many of the 
incentives for agency self-policing that exist in more thorough 
schemes of judicial review. The officials involved in adjudication 
would know that in some cases (exacdy which ones cannot be known 
in advance) the independent judicial branch will be reviewing their 
work. But the initial access to the courts would be of a stricdy lim- 
ited character. Within perhaps forty-five days, judicial review in a 
large majority of cases would be at an end, and the underlying 
deportation or exclusion order would become fully enforceable.'*^ 

G. Deportation 

When the asylum claim is finally denied, the underlying deporta- 
tion or exclusion order must be prompdy executed. Prompt execu- 
tion assures the only effective form of deterrent that does not 
depend on indiscriminate harshness meted out to all asylum seekers 
whatever the strength of their claim.'*** Surprisingly few such depor- 
tations occur at present, however, unless the alien has remained in 
detention.'*^ The reason is simple. Asylum seekers occupy a low 
priority for use of scarce investigation and enforcement resources in 
the district offices. Those resources are targeted instead on criminal 



*^ A possible complication might arise under the proposal offered here 
because deportability and asylum would be decided in different venues. The Justice 
Department should be able to overcome this complication, however. In arranging 
for unified judicial review, following deportation proceedings, of denials of 
legalization under INA §§ 210(e), 245A(0. 8 U.S.C. §§ 1160(e), 1255a(f) (1989). 
despite an initial splitting of adjudication forums between Legalization Offices and 
immigration judges, the Justice Department is gaining valuable experience in 
working out such technical details. See Manin.Judicial Retnew of Legalization Denials, 65 
I^^^ERPRETER Releases 757, 761 (1988). 

»o* See Martin, supra note 1, at 12-13. 

506 See supra notes 188-190 and accompanying text. 



352 DAVID A. MARTIN 

aliens and others apparently involved in major abuse of the immigra- 
tion system. 

If we were to look only at each individual category in isolation, 
this ranking of enforcement priorities makes sense. Criminal aliens 
do pose a greater threat to society than failed asylum applicants, who 
are largely harmless and law-abiding job-seekers. Nevertheless, 
enforcement priorities must be reoriented to consider more than just 
individual characteristics. They must also take full account of sys- 
temic impacts. The rest of this proposal painstakingly seeks every 
reasonable opportunity to streamline each stage of an inevitably 
complicated procedure. That effort is for naught — any achievements 
are rendered illusory — unless this last piece is resolutely inserted 
into the puzzle. 

H. Emergency Responses to Large-Scale Influxes 

This study was chartered at a time of relative stasis and calm 
within the asylum adjudication system. For years applications in the 
INS district offices had remained at an annual level of 20,000 to 
30,000. Although these figures ran some ten times higher than 
annual statistics in the mid-1970s, the number appeared politically 
tolerable. There was no undue pressure for quick fixes or emer- 
gency solutions. It would have been an auspicious time to provide 
for a phase-in of the ambitious changes suggested here, allowing for 
careful restructuring of offices and processes, the recruitment and 
training of new officers, and the inevitable adjustments and modifica- 
tions that will appear advisable as actual implementation reveals new 
problems and opportunities. 

But now, toward the conclusion of the project, the political situ- 
ation has altered considerably, and any changes will have to be 
implemented in much less favorable circumstances. Large influxes 
of Central Americans to Florida and Texas during the winter of 
1988-89 strained arrangements even for basic provision of shelter 
and food, "False refugees" claimed the front pages again, and the 
potential for political backlash reappeared. Radical solutions have 
been tendered, sometimes reflecting little understanding of the 
international and domestic legal framework.'*^^ Unfortunately, crisis 
is often necessary to generate the political will to make changes 
rather than to limp along under the old system. But patience for 

507 For example, an internal draft of draconian legislation, entitled the "Asylum 
Anti-Abuse Act of 1989," has been cirailated within INS. See 66 Interpreter 
RE1.EASES 478-79 (1989). 



REFORMING ASYLUM ADJUDICATION 353 

long-term solutions of the kind sketched here is in shortest supply at 
such times. And even if the political leadership remains committed 
to long term reform, political pressures may demand of the executive 
branch a showing of visible, immediate, and effective action that will 
stem the flow and dispatch pending cases quickly. 

I will therefore offer a few suggestions, plus a few words of cau- 
tion about some quick fixes that have been suggested. But above all, 
any crisis-driven requirement for prompt action must not divert 
attention fi-om the need to start implementing the central reforms 
proposed here as soon as possible. Almost every conceivable (and 
certainly every reasonable) emergency response will be easier to 
implement, to sustain, and to render effective, if emergency meas- 
ures are accompanied by the steady phase-in of a more reliable, high- 
quality, one-tier adjudication system staffed by a corps of true pro- 
fessionals, insulated from foreign policy concerns, sensitive to cross- 
cultural communication difficulties, and equipped to make effective 
use of the disparate array of information sources that must be 
employed. 

1 . Quick Denial of Manifesdy Unfounded Applications 

Several countries have made use of fast-track denials of "mani- 
fesdy unfounded" asylum claims (what I will call here "MU proce- 
dures"), and in time of large-scale influx, such possibilities become 
attractive. A well-designed MU procedure could conceivably help 
amplify the qualified deterrent message that the United States now is 
trying to send to those in Central America who may be contemplat- 
ing a trip northward.'**® But its contributions to this end would be 
modest, and its complications may outweigh its advantages, at least 
under current conditions. 

The UNHCR Executive Committee (a body in which the United 
States is a key participant) adopted a formal Conclusion in 1983 on 
"the problem of manifesdy unfounded or abusive applications" for 
asylum.'^ It gave cautious endorsement to the creadon of an expe- 

S08 T^is proposal is necessarily founded on an assumption that a large majority 
of the recent Central American asylum seekers do not qualify for asylum under 
current legal standards. That assumption is debatable, on legal and factual grounds, 
but for reasons that cannot be elaborated here, I believe it to be defensible. 

so« Conclusion No. 30 (XXXIV), 38 U.N. GAOR Supp. (No. 12A) at 25-26, U.N. 
Doc. A/38/12/Add.l (1983). Conclusion No. 30 states in relevant part that the 
UNHCR Executive Committee: 

(c) Noted that applications for refugee status by persons who clearly 
have no valid claim to be considered refugees under the relevant criteria 



354 DAVID A. MARTIN 

ditious procedure for dealing with such applications, but it warned 
against over-use of such measures, particularly in view of the "grave 
consequences" of an erroneous determination.* *° It therefore 
emphasized that any interview used in such procedures should be 
conducted by a "fiilly qualified official" and that the decision should 
be made by the "authority normally competent to determine refugee 
status."'** The Elxecutive Committee was primarily concerned that 
such responsibilities would devolve on border police,'*^ who would 
be ill-equipped to oury out the role and might have incentives to use 
MU procedures to exclude asylum seekers without an adequate effort 



consdtute a serious problem in a number of States parties to the 1951 
Convention and the 1967 Protocol. Such applications are burdensome to 
the affeaed countries and detrimental to the interests of those applicants 
who have good grounds for requesting recognition as refugees; 

(d) Considered that national procedures for the determination of 
refugee status may usefully include s(>ecial provision for dealing in an 
expeditious manner with applications which are considered to be so 
obviously without foundation as not to merit fiill examination at every 
level of the procedure. Such applications have been termed either 
"clearly abusive" or "manifesdy unfounded" and are to be defined as 
those which are clearly fraudulent or not related to the criteria for the 
granting of refugee status laid down in the 1951 United Nations 
Convention relating to the Status of Refugees nor to any other criteria 
justifying the granting of asylum; 

(e) Recognized the substantive character of a decision that an 
application for refugee status is manifesdy unfoimded or abusive, the 
grave consequences of an erroneous determination for the ^plicant and 
the resulting need for such a decision to be accompanied by appropriate 
procedural guarantees and therefore recommended that: 

(i) As in the case of all requests for the determination of refugee 

status or the granting of asylum, the applicant should be given a 

complete personal interview by a fully qualified ofRdal and, 

whenever possible, by an official of the authority competent to 

determine refugee status; 

(ii) The manifesdy unfounded or abusive charaaer of an application 

should be established by the authority normally competent to 

determine refugee status; 

(iii) An unsuccessful applicant should be enabled to have a negative 

decision reviewed before rejection at the frontier or forcible 

removal from the territory. Where arrangements for such a 

review do not exist. Governments should give favorable 

consideration to their establishment. This review possibility can 

be more simplified than that available in the case of rejected 

applications which are not considered manifestly unfounded or 

abusive. 



Id. 



»«o Set id. paras, (d). (e). at 25. 
'" Id paras. (e)(i). (e)(ii). at 25. 
^^^ See id para. (e)(iii), at 25. 



REFORMING ASYLUM ADJUDICATION 355 

to find the facts. Thus, such decision authority should be given to 
the system's ordinary asylum adjudicators. 

With these cautions in mind, MU procedures for the United 
States might be built on the following framework. Those persons 
apprehended by INS or "walking in" to INS offices who apply for 
asylum, or otherwise express fear of being returned to the home 
country, should be given the application form and told to complete 
and file it within a limited time period. Upon receipt of the form, an 
asylum adjudicator would go over the form and any other informa- 
tion in the alien's file to perform a preliminary screening. If the case 
clearly seems to have substance, it should simply be set for the regu- 
lar interview or hearing procedure. If it looks as though the claim 
might be manifestly unfounded, it should be set for an early MU 
screening interview, which should be recorded verbatim. 

If, during the MU screening interview, the applicant tenders a 
plausible basis for her asylum claim, the matter should be passed on 
to the next stage, the merits hearing, which would probably take 
place several weeks later. Access to the full merits hearing should be 
permitted even if the applicant's present account seems to contradict 
earlier statements g^ven to the immigration officials. The individual 
may be pressed about seeming contradictions, but, unless the 
responses reveal dear and continuing fi-aud, the applicant should 
make it to the next stage in the process.'*' There are simply too 
many possible inno cent explanations for inconsistent initial state- 
ments in these settings, owing to the manifest difficulties of cross- 
cultural communication and to the understandable reticence that 
truly persecuted people may feel upon their first encounter with uni- 
formed American officials. 

The MU procedure is not the forum for resolving such contra- 
dictions. To do so adequately would require expanding the proce- 
dure until it became virtually indistinguishable fi-om the merits 
procedures."* This fact unavoidably limits the utility of MU proce- 



'" Apparently a high percentage of the current applicants say something about 
coming to the United States for a job during their first encounters with INS, and only 
later begin speaking of feared persecution. Because of cross-cultural differences, one 
cannot simply apply a presumption that the first statement is the more accurate or 
honest (even if such a presumption might make sense in dealing with American 
citizens in other contexts). It is entirely possible that the individual muttered a non- 
threatening response in the first encounter with uniformed officials only because her 
entire experience in her home country taught her to volunteer nothing to people in 
uniforms. Tlie change of story, of course, should be explored fully in the merits 
hearing, but it cannot be treated as dispositive in the MU procedure. 

*•* The Department of Justice actually drafted a kind of MU procedure, using 



356 DAVID A. MARTIN 

dures. If the individuals come from countries with known human 
rights problems (and this includes the Nicaraguans, Salvadorans, and 
Guatemalans who make up the bulk of the current influx), MU proce- 
dures will probably screen out only a handful of them. This handful 
would consist mainly of persons so poorly advised by friends or 
"travel agents"— or so honest — that they speak during the MU inter- 
view only about crop failures at home and job attractions here. But a 
more restrictive approach carries too high a risk of quick return of 
true refugees. 

If the application is found to be manifesdy unfounded, the con- 
sequence would be some truncation of normal procedures. Again, 
many variants are possible; the most effective would require statu- 
tory change. Possible limitation by regulation alone, however, might 
take the following shape: The MU determination should constitute a 
final negative ruling on the asylum and withholding claims, without 
the possibility of de novo consideration or further review in any 
administrative forum.'** 

If statutory changes are deemed advisable, the MU determina- 
tion could also serve to limit judicial review, but with safeguards. 
Limited judicial review would be possible, through a summary proce- 
dure like that suggested above for deciding on "leave to appeal." 
There would be one difference. If the court found the MU determi- 
nation unsupported, it would not schedule fiill-fledged court review. 
It would simply remand the case for a full merits hearing before the 

different terminology, in asylum regiilations proposed in 1978. Those regulations 
allowed for "summary judgment" on a shortened timetable, in cases that appeared 
straightforward — i.e., that appeared to lack any genuine issues of material fact. See 43 
Fed. Reg. 40.801 (1978). The 1978 proposal never made it into fiilly operative final 
regulations, however. The agency explained its abandonment of the idea as follows: 

Critics cite as hardships [rendering the summary judgment procedure 
unfair] the circumstances of many newly arrived applicants for asylum 
who face a langiiage barrier and suffer fi-om a lack of education and 
limited financial resources .... 

The objections to the summary judgment provisions have merit. 
Upon reconsideration, this type of motion appears to be rather 
sophisticated given the nature of the proceedings and situation of the 
individuals .... [Also, s]ince applications for asylum most frequendy 
involve disputed facts a motion for summary judgment would rarely lie. 

44 Fed. Reg. 21.254 (1979). These considerations have lost htde force in intervening 
years and may still counsel against adopting such a procedure, under whatever name. 
"5 An exception might be made for review solely on the initiative of the 
Chairperson of the Asylum Board, much as the Chief Administrative Hearing Officer 
may now arrange for review of certain administrative orders issued by Administrative 
Law Judges under 8 U.S.C. §5 1324a. 1324b (1988) within Uiirty days of the order's 
issuance. 5** 28 C.F.R. { 68.52 (1988). 



REFORMING ASYLUM ADJUDICATION 357 

adjudicator. On the other hand, if the court approved the MU find- 
ing, the deportation order would become final and immediately 
enforceable, without the possibility for further judicial consideration. 
Obviously the force given an MU determination cautions that 
MU procedures should not be used unless training and recruitment 
have proceeded to the stage that the Justice Department has substan- 
tial confidence in the officers doing the MU interviews. The ever- 
present temptation will be to overuse MU determinations. A report 
prepared by Professor A.M.J. Swart of the Netherlands for the Coun- 
cil of Europe found that exacdy this sort of error was being commit- 
ted. She reported that national authorities implementing MU 
procedures in several European countries 

are inclined to want to judge the merits of a request fiilly in order 
to see whether it is abusive or unfounded [rather than 'dearly* abu- 
sive or 'manifestly' unfounded]. This means that criteria which 
have been developed to do no more than make a first, rough selec- 
tion possible, become so important that the selection itself 
becomes the crucial moment in asylum procedure for all asylum 
seekers.'*^ 

The lack of adequately trained and equipped personnel at pres- 
ent may pose a substantial obstacle to effective implementation of 
MU procedures as an immediate response to the current high appli- 
cation rate. Such a procedure would be far more reliable once a staff 
of independent and professional asylum adjudicators is in place — 
another reason to move quickly toward implementation of such a 
new administrative scheme. Still, MU procedures should not be 
expected to carry a heavy load. Even Germany, which has had sev- 
eral years to perfect its MU techniques, can dispose of only about 
twenty-five percent of its cases in this fashion, and Canada has elimi- 
nated only about ten percent this way under its new system.'" A 
better solution, as the UNHCR's formal Conclusion on this issue 

S'6 Swart, The Problems Connected With the Admiision of Asylum Seekers to the Territory 
of Member States, in The Law of Asylum and Refugees: Present Tendencies and 
Future Perspectives 65, 80 (Council of Europe 1987). 

**' See von Pollcm, Die Entwicklung der Asylbewerberzahlen wn Jahre 1988, 1989 

ZErrSCHRlFT FUR AUSLANDERRECHT UND AUSLANDERPOUTIK (ZAR) 23, 26 (from 1982, 

when MU procedures were introduced in Germany, through 1988, 25.32 % of cases 
were denied as "manifesdy unfounded;" for 1988, the figure was 26.38 percent); 
Refugee Determination in Canada: First Quarter Review 10 (Apr. 25, 1989) (noting 
that 89% of 2,037 claims concluded at the initial suge were referred for a fiill merits 
determination). A UNHCR study estimated that manifesdy unfounded or abusive 
claims constituted 10 to 15 % of "unscheduled arrivals" in industrialized countries in 
1984. S« Jaeger. Irregular Movements: The Concept and Possible Solutions, in The New 
Asylum Seekers: Refucess Law in the 1980s 23, 31 (D. Martin ed. 1988). 



358 DAVID A. MARTIN 

ultimately suggests, may be instead to target resources towards 
assuring speedy completion of full merits hearings and all review 
stages.*'® 

2. Disqualification for Transit Through Third Countries 

Some interest has also been expressed in new rules that might 
disqualify applicants fi'om asylum or withholding of deportation if 
they have passed through other countries where they could have 
applied for asylum before reaching the United States. Because virtu- 
ally all asylum seekers from Centr^J America travel through Mexico, 
some regard this as a nifty device to deal with most of the current 
caseload. 

This proposal, however, poses such serious legal and practical 
problems that it should be abandoned.' '^ Suppose a Salvadoran 
files for asylum and is told that her claim will not be heard on the 
merits because she should have applied in Mexico. What exactly will 
be done with her? Presumably she could be sent to Mexico, if Mex- 
ico would agree to receive her, but the odds that Mexico would 
accept such a person, much less tens of thousands of needy 
Salvadorans, are almost nonexistent.'**' The only country likely to 
accept her would be El Salvador, the country where she claims she 
would be persecuted. Her transit through Mexico to the United 
States by no means proves that she had no legitimate fears in El Sal- 

»>8 In Conclusion No. SO. the UNHCR Executive Committee: 

(f) Recognized that while measures to deal mth manifestly 
unfounded or abusive applications may not resolve the wider problem of 
large numbers of applications for refugee status, both problems can be 
mitigated by overall arrangements for speeding up refugee status 
determination procedures, for example by: 

(i) Allocating sufficient personnel and resources to refugee status 
determination bodies so as to enable them to accomplish their 
task expeditiously, and 
(ii) The introduction of measures that would reduce the time 
required for the completion of the appeals process. 
Conclusion No. SO, supra note S09, para, f, at 26. 

"® Similar issues have been debated in Euro|>e for years under the rubric of the 
"country of first asylum" doctrine, and the debate there suggests the legal and 
political intricacies that can be implicated. See generally Vierdag, The Qnmtry of "First 
Asylum": Some European Aspects, in The New Asylum Seekers: Refugee Law in the 
1980s 73 (D. Martin ed. 1988); Conclusion No. 15(XXX), Refugees Without a Qnmtry of 
Asylum. S4 U.N. GAOR Supp. (No. 12A) at 17, U.N. Doc. A/34/12/Add.l (1979). 

'*o See generally J. Friedland Se J. Rodriguez v Rodriguez, Seeking Safe 
Ground: The Legal SrruATioN of Central American Refugees in Mexico 55 
(1987): P. Fagen & S. Aguavo, Central Americans in Mexico and the Unfteo 
States: Unilateral, Bilateral, and Regional Perspectives (1988). 



REFORMING ASYLUM ADJUDICATION 359 

vador. Both Article 33 of the 1951 Refugee Convention and INA 
§ 243(h) obligate the United States not to return her to El Salvador if 
her fears are well-founded. 

In the end, therefore, a "transit" doctrine would not obviate a 
ruling on the merits, at least with respect to the nonrefoulement obliga- 
tion. It would only delay such a ruling and, in the meantime, possi- 
bly complicate diplomatic relations with Mexico. 

3. Ending Work Authorizations and Making Alternative 
Arrangements for Subsistence Pending Adjudication 

Government policymakers may yet perceive a need for some 
decisive step to send a deterrent message at times of sudden influx. 
Many INS personnel interviewed for this study volunteered a ready 
solution along these lines: simply end the work authorizations that 
are now. fairly automatic for asylum seekers during the pendency of 
their claims (both initially and on appeal).'^' Additional evidence 
also supports the theory that the recent rise in filings is at least par- 
tially linked to the work authorization issue.'^^ 

A simple end to work authorizations, however, or a raising of the 
threshold to qusdify beyond the "nonfrivolous" standard now con- 
tained in the regulations,'^' will neither fiilly solve the problem nor 
-likely be sustained by the courts, unless other steps are also taken. 
Before the rules were amended in 1987 to make work authorization 
nearly automatic, district directors had considerable discretion in 
granting such permission. In Diaz v. INS, however, a district court 
issued a preliminary injunction against restrictive implementation, 
finding that a restrictive policy unduly burdened the alien's statutory 
right to apply for asylum and thereby fhistrated the goals of the stat- 
ute.'** The nondiscretionary 1987 regulations were issued at least in 
part to conform to Diaz. '*' 

Nothing has happened since then to undercut the court's rea- 

"1 See 8 C.F.R. 8§ 274a.l2(c)(8) (1989) (providing for employment 
authorizations in increments not exceeding one year for any alien who has filed a 
non*frivolous application); 8 C.F.R. $ 274a.l3(d) (1989) (requiring adjudication of 
the employment authorization application within 60 days, the failure of which will 
result in a grant of interim employment not to exceed 120 days). An INS 
memorandum elaborates on the standards used to determine frivolousness. See 
Memorandum from Richard £. Norton, sufn-a note 146. 

522 See. e.g.. B. Frelick. supra note 214, at 17. 

923 5^^ 55 Interpreter Releases 4 (1989); Memorandum from Richard E. 
Norton, supra note 146. 

ss* See Diaz v. INS, 648 F. Supp. 638, 651-53, 656 (E.D. Cal. 1986). 

925 See Memorandum from Ignatius Bau and Robert Rubin, supra note 215, app. 



360 DAVID A. MARTIN 

soning. It is indeed quite plausible to read Congress's enactment of 
§ 208 in 1980 as creating a right for persons physically present in the 
United States to have their asylum claims heard on the merits. If 
work authorization is now to be denied, any lawyer for the Depart- 
ment of Justice is bound to be asked in court how the government 
expects asylum seekers to survive during the months (and possibly 
years) until a final ruling is obtained on the application. Unless the 
government takes further steps to provide for such people during 
the pendency of the claim, the lawyer has no respectable answer.'^® 
Courts might easily conclude that the government was trying to 
starve people out of pursuing a congressionally mandated right. 
And they would surely point out that a no-work-authorization policy 
falls as heavily on bona fide refugees as on the abusers who are the 
ostensible targets.'^^ 

At times of heavy influx, a policy of near-automatic work autho- 
rizations may well be ended, but the government must then provide 
alternative arrangements for feeding and housing the asylum seek- 
ers. Obviously, this course would be expensive, but it could be 
adopted in the hope that it would slow new arrivals and minimize 
overall long-term costs. If such a course is chosen, these alternatives 
could be provided in two ways. The first would be to detain asylum 
seekers physically under the relevant portions of the immigration 
statutes. Current regulations already provide for a presumption of 
detention for excludable aliens who arrive without documents.'*® 
The bonding provisions governing deportation probably also allow 
the Attorney General sufficient discretion, particularly during what 
could plausibly be argued are emergency conditions, to arrange for 



n at 882-86 (explanation of the significance of the 1987 regulations by the attorneys 
involved in Diaz). 

326 j)ie problem is compounded because asylum seekers are not considered to 
be "permanendy residing under color of law" (PRUCOL), which is a prerequisite to 
qualifying for most federally funded public benefit programs. See 20 C.F.R. 
§416.1618 (1989); see also Sudomir v. McMahon. 767 F.2d 1456. 1466 (9th Cir. 
1985); Wheeler. Alien Eligibility for Publtc Benefits: Part I, Immicr. BRiEnNCS. Nov. 
1988, at 3-4; Stein ie. Zanowic, Pemutnent Resident Alien Under Color of Law: The Opening 
Door to AUen Entitlement Eligibility. 1 Geo. Immicr. L.J. 2S1 (1986). These federal 
statutory restrictions sharply distinguish U.S. practices from those of most European 
countries, which routinely provide subsistence allowances and other benefits for 
asylum seekers within the general schemes they have for public assistance. This 
difference helps explain European countries' more ready resort to denials of work 
authorization to asylum applicants, even though such denials clearly impose a larger 
burden on the taxpayers. 

527 See Duo, 648 F. Supp. at 655-56. 

528 See 8 C.F.R. fi§ 212.5(b). 235.3(b) (1989). 



REFORMING ASYLUM ADJUDICATION 361 

detention of a large proportion of those asylum seekers who have 
already made an entry into the United States. 

The second framework would emphasize the voluntary nature of 
the communal shelter and feeding facilities. It would be set up pri- 
marily to assure that no asylum seekers go hungry while awaiting a 
ruling on their asylum claims. If they cannot provide for themselves 
through personal resources or the resources of friends and family, 
asylum seekers could move into the governmental facilities. For 
those who choose this course, some kind of daily check-in procedure 
might be used to verify identities and to maximize the chances that 
individuals can be located when a deportation order becomes final. 
Presumably, they could come and go at their discretion during the 
day. (West Germany uses such arrangements in the communal hous- 
ing facilities it has established for asylum seekers.) This course of 
action would probably cost less in direct government payouts 
because a fair number of asylum seekers would prefer to move in 
with their families. Of course, many of those not in the government 
facility might well attempt to work surreptitiously or with false docu- 
ments. Moreover, this course of action would probably also increase 
the absconding rate once final deportation orders begin to issue. 

Under either course, there will obviously be a need to locate 
considerable government facilities for housing, but there is a well- 
worn path of experience here, extending back to the 1975 refugee 
emergency caused by the fall of Saigon and the 1980 Mariel boatlift. 
The current population of asylum seekers should, of course, be con- 
siderably easier to deal with than the Mariel population, which 
included some inmates fi-esh from Cuban jails. Steps should be 
taken to make the new focilities as comfortable as possible under the 
circumstances, and to minimize some of the pathologies that are gen- 
erated when enforced idleness and close quarters continue for 
lengthy periods. For example, it would be advisable to keep families 
together and to provide access, whenever possible, to cooking facili- 
ties, so that the individuals could prepare their own meals. The 
detainees are not felons, and the government might gamer wider 
public support for any such deterrence policy if it attempts to ame- 
liorate camp conditions as much as possible.^^^ 

S29 yhis policy appears consistent with what the Select Commission on 
Immigration and Refugee Policy had in mind when it proposed creation of asylum 
"processing centers." See Select Commission, suffra note 69, at 166-68. Moreover, 
the government gained some further experience in establishing facilities of this sort 
in South Texas in early 1989, as part of its response to the sharp increase in asylum 
seekers arriving there. While the Department of Justice detained many of these 



362 DAVID A. MARTIN 

Of course, one possible objection is that any steps to ameliorate 
conditions limit the deterrent impact. That risk is worth taking. The 
main deterrents will remain denial of work authorization (which 
should be widely publicized through available media in the region of 
origin), and new measures to hasten final decisions, which would 
accelerate both the return of the unqualified applicants to their 
homelands and the release of those who merit asylum. 

In the end, the difhculties and disadvantages of any such policy 
should not be minimized. Outright detention clashes glaringly with 
the proud heritage symbolized by the Statue of Liberty. Less harsh 
communal housing, at public expense, is apparendy hard to sustain 
politically, for it can be portrayed as a pure welfare program for ille- 
gal immigrants, fed and sheltered entirely at taxpayer expense. 
Either course makes sense only as a temporary palliative; it would be 
much better to avoid them altogether. Avoidance is possible, in the 
long run, if an effective and expeditious adjudication system, as 
sketched above, can be patiendy implemented and sustained. That 
sort of system allows for deterrence through the prompt deportation 
of the unqualified, rather than through enforced idleness (at mount- 
ing governmental expense) while adjudication grinds slowly onward. 

V. Conclusion 

Government officials reading through all the proposals offered 
here may be struck by the apparent cost of the system envisioned, 
compared to what we seem to have today. Until now, asylum respon- 
sibilities have been assumed by a mere handfiil of harried examiners 
in district offices and by the surprisingly small corps of immig^tion 
judges who shoehorn asylum in among their many other responsibil- 
ities. The true costs of the present system, however, are much 
higher. They include not only the costs to localities in Florida, 
Texas, and California that are scrambling to meet the elemental 
needs of asylum seekers now applying in much higher numbers, but 
also the costs that are likely to accrue in the future as the magpnet 
effect compounds."** Absent effective adjudication reforms, we are 

p>eople in jail-like facilities, it also set up a more open Alien Shelter Care Program, 
available mainly to families and unaccompanied minors. See Justice Department 
Announces Funding for Services for South Texas Aliens, 66 Interpreter Releases 720 
(1989). 

'*o See Appendix. See generally General Accounting Office Letter to Sen. Bob 
Graham (Feb. 23. 1989) (GAO/GGD-89-54FS) (recounting financial effects of asylum 
applicants on Miami); Schmalz, Miami Pupils: Future Hope is Crisis Now, N.Y. Times, 
Mar. 21, 1989, at Al, col. 2; McCoy. Christie & de Cordoba. In Mumi, Rivalries of 



REFORMING ASYLUM ADJUDICATION 363 

unlikely to escape an expensive detention or government accommo- 
dation scheme whenever the flow reaches a high level, as the govern- 
ment's response to the spike in applications during the winter of 
1988-89 demonstrated."' 

The costs required to implement the reformed syslfem are 
worthwhile if the changes can break the vicious circle in which asy- 
lum policy now seems to be caught. Quicker, seemingly cheaper 
fixes are wholly illusory. They were tried in the Haitian Program of 
1978. The result was only years of litigation, preliminary injunc- 
tions, remands, and duplicative reconsideration, topped off by a 
major award of attorneys' fees to the asylum seekers' counsel,'*^ 
The courts have repeatedly shown that they will intervene unless the 
asylum "problem" is addressed by a comprehensive program that 
demonstrates adequate seriousness about our Statue of Liberty tra- 
dition. Such seriousness inevitably costs money. 

Refugee advocates encountering these proposals will probably 
be struck instead by the possible removal of several layers of com- 
forting checks and appeals. Those checks have probably been effec- 
tive in assuring that bona fide refugees are not sent home, 
particularly if a skilled advocate makes full use of all possible avenues 
of attack. But the cost has been high. It has meant the creation of a 
system that has great difficulty actually sending anyone home. Now 
that this latter message has been received in Central America (and to 
some extent all over the globe), the flow will probably continue to 
rise, until political backlash imposes its own correctives— correctives 
likely to be far more draconian. The effort here is to find ways to 
minimize the magnet effect without impairing the quality of the judg- 
ment on the merits of the asylum claim. Indeed, the steps proposed 
here, if properly implemented and carefully monitored, should sig- 
nificantly improve the accuracy and fairness of decisionmaking, 
despite the streamlining of the system and the trimming of layers of 
review. 

This proposed system, centered on a nonadversarial model of 
adjudication, obviously places great reliance on the role of the single 
adjudicator. One refugee lawyer, apprised of an early version of 



Mtnority Groups Fan Riot's Flames, Wall St. J., Jan. 20, 1989, at 1, col. 4; Thompson, 
Black Anger at Refugees Fed Riots. Wash. Post, Jan. 20, 1989, at A3, col. 1; Applebome, 
sufrra note 4. 

"' In February 1990, in response to a new increase in applications, INS 
announced the resumption of more systematic detention of asylum applicants. See 
Suro, U.S. Is Renewing Border Detentions, N.Y. Times, Feb. 8, 1990, at A22. col. 1. 

"2 5^ si^a note 3. 



364 DAVID A. MARTIN 

these proposals, marveled: "You would really have to have trust" in 
the officials running the procedure. Exactly. Asylum adjudications 
have been conducted for years in such an atmosphere of profound 
and mutual mistrust that we may have great difficulty even conceptu- 
alizing such a system. But if we are to show true fidelity to the best of 
our asylum tradition, we have to find a way to create a system that 
merits our confidence. Other systems are too cumbersome to work 
effectively — at least when the asylum applicant intake reaches 
100,000 per year. 

Asylum determinations should be made by carefully chosen and 
well-trained specialists who possess focused responsibility for fulfil- 
ling our legal obligations and for implementing consistent, coherent, 
and accurate policy. Courts must develop a more deferential stance 
toward that expertise. It is time to create a system that will, at long 
last, merit such deference and trust, even on issues that wiU remain 
hotly controversial and about which we, as Americans, rightly care 
deeply. 



REFORMING ASYLUM ADJUDICATION 365 



Appendix 

ASYLUM ADJUDICATION REFORM: COST ESTIMATES 
(as of June 1989) 



I. Costs of the Present Asylum Adjudication System 

A. Adjudication Costs (in millions of dollars) 

Elxecutive Office of Immigration Review (EOIR) $10.0 

(Includes immigration judges and 
Board of Immigration Appeals) 

INS: Trial attorneys and support staff 4.0 

District offices 3.0 

Central office 0.5 

Asylum Policy and Review Unit, Department of 0.5 

Justice 

Asylum office. Department of State 0.5 

TOTAL $18.5 

Financial data for the various units of the Justice Department 
and State Department are not kept in a manner that readily permits 
identification of those costs associated with asylum adjudication. 
These figures therefore set forth rough, but probably conservative, 
estimates for costs attributable to asylum adjudication for FY 1988, 
when the system was equipped to handle approximately 20,000- 
30,000 applications, based upon telephone interviews with officials 
in each affected department. 

Note: TTiis list makes no provision for detention, deportation, Coast 
Guard interdiction, assistance to persons eventually granted asylum, 
or state and local government expenditures. 

B. Detention 

Detention costs $30-35 per person per day. Therefore deten- 
tion of 4500 individuals for one week costs the federal government 
$1 million. During May 1989, INS was detaining an average of about 
3700 asylum seekers in South Texas alone.' Other areas of heavy 
concentration of asylum seekers (notably metropolitan Miami and 

• See Refugee Reports, May 19. 1989, at 8. 



366 DAVID A. MARTIN 

southern California) also give rise to substantial numbers of 
detainees. 

C. State and Local Government Costs 

The arrivjd of large numbers of asylum seekers in a concentrated 
period can have a considerable effect on local government expendi- 
tures for which there is no readily available federal reimbursement. 
No overall figures are available, but a GAO report examining the 
effect on the Miami-Dade County area of recendy arriving 
Nicaraguans, whb came in large numbers beginning in mid- 1988, 
provides an iUustrative picture of such local costs. ^ 

GAO reported an estimated $4 million in social services pro- 
vided by Dade County to Nicaraguan asylum seekers in calendar 
1988 (averaging over $10,000 per day). This figure does not include 
educational expenses, which were expected to be high. In addition, 
the city of Miami reported spending $4,000 a day on recent Nicara- 
guan arrivals. Thus, not counting educational costs, local govern- 
ment in the Miami area incurred expenses of over $14,000 per day 
for Nicaraguan asylum seekers during the 1988-89 influx.* 

II. Estimated Costs for Reformed Adjudication System 

Because the recommendation proposes that asylum adjudicators 
be equivalent in rank and salary to immigration judges, the following 
estimates for adjudicators are based on figures provided by EOIR for 
full costs of additional immigration judges and support personnel. It 
is estimated that thirty adjudicators would be needed to handle each 
increment of 10,000 asylum decisions'* per year (333 cases per adju- 
dicator per year or approximately seven cases per week). 

2 General Accounting Office, Political Asylum Appucants: Financial 
Effect on Local Services in the Ml<vmi Area (1989). 

' This GAO Import sets forth costs only for Nicaragiums in the Miami area, 
although Haitian and other asylum-seekers had been arriving in increasing numbers 
f there. (The Miami district office of INS then ranked second in asylum application 
receipts among distria offices, receiving 8.214 asylum applications in FY 1988 and an 
estimated 15,(X)0 in 1989; the Los Angeles office was first with 28,491 applications in 
FY 1988 and an estimated S4,0(X) in 1989.) 

^ Roughly 10 to 20 percent of asylum applications are withdrawn or abandoned 
before decision. Therefore 10.000 decuwns should keep pace with 1 1.000 to 12.500 
(plications. 



REFORMING ASYLUM ADJUDICATION 367 

Adjudication Costs (in millions of dollars) (for 30,000 
decisions per year) 

General administration and program direction $1.0 

Adjudication division 

Adjudicators 9.0 

Support staff for adjudicators 7.5 

Appellate division 

Board members and immediate staff 0.5 

Staff attorneys and other support staff 4.5 

Documentation center 1.0 

TOTAL $23.5 

The estimate assumes a capacity of 30,000 decisions per year, to 
show costs that would be comparable to the upper estimate of capac- 
ity for the current system as it operated in FY 1988. If the lower 
estimate were used instead for comparison (a caseload of 20,000 
decisions), the reformed system would cost $16.5 million. 

Actual adjudication output requirements will surely be higher, at 
least in the early years, given the 1989 appUcation rate. Once the 
new system is fully operational, however, it should reduce the 
number of asylum applications and diminish or eliminate the need 
for detention as a deterrent, thus saving substantial federal 
expenditures. 



To SS 9-5-89 



m 



WHICH AGENCY INTERPRETATIONS SHOULD BIND 
CITIZENS AND THE COURTS? 



by 

Robert A. Anthony* 



*B.A. 1953, Yale University; B.A. Juris. 1955, M.A. , Oxford m 

University; J.D. 1957, Stanford University. Professor of Law, 

George Mason University. Chairman of the Administrative Hi 

Conference of the United States 1974-1979. ''"' 

This Article is adapted from a report prepared tor the m 

Administrative Conference of the United States, which formed the "|| 
basis for the Conference's Recommendation 89-5. The views herein ' 
are those of the author, and should not be attributed to the 
Administrative Conference. 

I owe special gratitude to Peter Strauss and Ronald Levin 
for valuable consultations during the development of this 

Article. My warm thanks for their insightful comments on drafts {j||| 
go also to Joan Caton Anthony, Carl A. Auerbach, Stephen G. 
Breyer, Clark Byse, Kenneth Culp Davis, Walter Gellhorn, Sheldon 
E. Hochberg, Charles H. Koch, Jr., Jeffrey S. Lubbers, Richard J. 
Pierce, Jr., Kenneth W. Starr, Thomas M. Susman, Don Wallace, I'li' 

Jr., Stephen F. Williams, and Katherine Yarbrough. All 
responsibility, of course, is mine. 



m 
iiilii 



AGENCY STATUTORY INTERPRETATIONS 371 

TABLE OF CONTENTS 

Page 
Introduction and Summary 

I . Prior Patterns of Acceptance and Deference 

A. Legislative Regulations 

B. Interpretations Expressed in Adjudicatory Decisions. . 

1. Adjudication of Mixed Questions 

2. Adjudication of Pure Questions and Major 

Interpretive Questions 

3 . Summary 

C. Other Inteirpretive Formats 

D. Towards Clarity in the Pre-Chevron Law 

II . The Chevron Doctrine 

A. Chevron's "Step 1" 

1. Interpreting for Conformity with Specific Intent 

2 . "Pure Questions" of Interpretation 

3. Interpreting for Violations of Statute 

B. Chevron's "Step 2" 

1. Judicial Acceptance of Reasonable Agency 

Interpretations 

2. The Place of the Delegation Inquiry in Step 2... 

3. The Key Inquiry: Is the Interpretation in This 

Form Binding? 

4 . A Suitable Standard of Review 

V. Which Interpretive Formats Should Bind? 

A. Legislative Rules 

B. Agency Adjudications 

1 . Mixed Questions 



372 ROBERT A. ANTHONY 

TABLE OF CONTENTS (continued) 

Page 

2 . Pure Questions 

3 . Reasoned Decisions 

4 . Lower-Level Decisions 

C. Procedural Rules and Actions 

D. Interpretive Rules and Policy Statements 

E. Manuals, Guidelines, Staff Instructions, 

Opinion Letters 

F. Litigating Positions 

G . Miscellaneous Formats 

Conclus ion 



AGENCY STATUTORY INTERPRETATIONS 373 



I. INTRODUCTION AND SUMMARY 

Agencies continually interpret the statutes they administer. 
Their interpretations are expressed in a great variety of 
forms — in legislative regulations, adjudicatory opinions, 
manuals, court briefs, interpretive rules, policy statements, 
staff instructions, opinion letters, audits, correspondence, 
informal advice, guidelines, press releases, testimony before 
Congress, internal memoranda, speeches, explanatory statements in 
the Federal Register, and others. 

This Article considers the extent to which these varied 
forms of agency interpretive expression, sometimes referred to as 
"formats," should be accepted or accorded some sort of 
consideration by the courts that review them. 

This inguiry is part of a larger one, which forms its 
context: Under what circumstances should it be the agencies, 
rathei than the courts, that place definitive interpretations 
upon federal statutes? Put another way, when is a reviewing 
court bound to accept the agency's interpretation? 

Interpretation of a statute presents a question of law, and 
"[i]t is emphatically the province and duty of the judicial 
department to say what the law is."' "[0]ne of the judiciary's 
characteristic roles is to interpret statutes * * * . " But in 
reviewing interpretations placed upon statutes by the agencies 
that administer them, the courts for many decades have accorded 
respectful attention or even controlling effect to such agency 
interpretations . ^ 

In some situations, the court deems itself bound to accept 
the agency's interpretation outright, thereby giving it 
controlling effect, provided only that it is consistent with 
statute and is reasonable. In other cases, the court reserves 



Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803) 
(per Marshall, C.J.). 

Japan Whaling Ass ' n v. American Cetacean Soc'y, 106 
S.Ct. 2860, 2866 (1986). 

"[C]ourts do not necessarily abdicate a Marshallian 
duty to 'say what the law is* by deferring to agencies. Courts 
retain the authority to control administrative abuses of power; 
deferential review simply recasts the question of law as whether 
the agency's interpretation is 'reasonable'." Diver, Statutory 
Interpretation in the Administrative State . 13 3 U. Pa. L. Rev. 
549, 569 (1988), quoting Monaghan, Marbury and the Administrative 
State, 83 Colum. L. Rev. 1, 27-28 (1983). See also Levin, 
Identifying Questions of Law in Administrative Law , 74 Geo. L. J. 
1, 21 (1985). 



374 ROBERT A ANTHONY 



the power to arrive independently at its own interpretation; it 
should give respectful consideration to the agency's 
construction, but may reject it, even if it seems a reasonable 
one. 

The law governing judicial acceptance of agencies' 
interpretations of the statutes they administer is now dominated 
by Chevron U.S.A. Inc. v. Natural Resources Defense Council. 
Inc. This case has been cited by the Supreme Court dozens of 
times, and by the lower federal courts, hundreds. Chevron is 
understood to require a reviewing court to accept an agency 
interpretation that (a) is not contrary to statute or specific 
statutory intent and (b) is reasonable. (I shall refer to this 
judicial process as "Chevron acceptance" of the agency 
interpretation. ) 

This understanding unquestionably governs the usual case in 
which the agency, in pursuance of delegated authority, issues its 
interpretation through a legislative regulation or other agency 
action possessing the force of law.* But what if the same 
interpretation — identical in its substantive content — were 



Cases and commentary have confusingly used the term 
"deference" to refer to both of these approaches, even though 
they imply different roles for the reviewing court. Indeed, the 
courts often have been quite unclear about which approach they 
were following. For precision, I will refer to the first 
approach as "acceptance", or sometimes in more specific contexts 
as " Chevron acceptance" or " Hearst acceptance." See Chevron 
U.S.A., Inc. V. Natural Resources Defense Council, Inc., 467 U.S. 
837 (1984); NLRB v. Hearst Publications, Inc., 322 U.S. Ill 
(1944). 1 shall call the second approach "special consideration," 
or more specifically " Skidmore consideration." See Skidmore v. 
Swift & Co., 232 U.S. 134 (1940). Under the analysis presented 
in this Article, there can be a third approach. In cases of pre- 
enforcement review, where the court does not need to make a final 
and complete interpretation, it may limit its review to the 
question of whether the agency's interpretation is demonstrably 
wrong, without deciding whether it is right or acceptable. See 
text at notes 182-183, infra. 

^ 467 U.S. 837 (1984) (6-0 decision; opinion by Stevens, 
J.; Marshall, Rehnquist and O'Connor, JJ. not participating). 

' An agency interpretation "has the force of law", and 
therefore is "binding" upon the courts (as well as upon the 
public and upon the agency itself) , when a court may not review 
it freely, but must accept it unless it is contrary to statute or 
unreasonable. A legislative regulation issued pursuant to 
delegated statutory authority is an example of agency action 
possessing these characteristics. See Batterton v. Francis, 432 
U.S. 416, 425-26 (1977) quoted at note 17 infra . See also 
Chrysler Corp. v. Brown, 441 U.S. 281, 301-04 (1979). 



AGENCY STATUTORY INTERPRETATIONS 375 



expressed only in an opinion letter, a policy statement, a press 
release, or an amicus brief? 

It would be the rare case in which the agency possessed 
congressionally delegated authority to make definitive 
interpretations, carrying the force of law, through such informal 
issuances. Nevertheless a reviewing court — under a broad and 
perhaps uncritical reading of Chevron — might feel compelled to 
accept an interpretation of this type, without regard for the 
agency's lack of delegated authority to issue authoritative 
interpretations through such informal formats. A number of 
courts have done so. 

Such a reading of Chevron is possible because its language 
suggests that, if the reviewing court cannot find specific 
congressional intent on the precise point at issue, but rather 
finds the statute to be silent or ambiguous on that point, it 
should presume that Congress delegated to the agency the 
interpretive authority to fill the gap, and must therefore accept 
any reasonable agency interpretation. It is then but a short 
step, perhaps unconsciously taken, to assume that this implied 
delegation requires acceptance of the interpretation without 
regard for the format in which it is expressed. 

Should the courts under Chevron be bound to accept (and not 
merely to consider) reasonable agency interpretations expressed 
in informal formats? The present study addresses this question, 
and yields generally negative answers. Where the format is an 
informal one, it ordinarily does not carry the force of law, and 
a reviewing court is not bound by the agency interpretation, 
though it should give special consideration to the agency 
opinion. 

The threshold issue for the court is always one of 
congressional intent: did Congress intend the agency's 
interpretation to bind the courts? The touchstone in every case 
is whether Congress intended to delegate to the agency the power 
to interpret with the force of law in the particular format that 
was used . 

Sometimes the reviewing court will conclude that Congress 
has mandated or prohibited a specific substantive position. In 
such a case, obviously, the court cannot find that Congress 
intended a delegation to the agency co interpret inconsistently, 
and an agency interpretation may enjoy the force of law if it 
conforms to the substantive congressional intent. In every other 
case, however, an agency interpretation that claims the force of 
law must be grounded in an adequate delegation. 

When looking for delegation, the reviewing court must of 
course first satisfy itself as to the agency's delegated 
authority to make binding interpretations with respect to the 
subject matter . Under Chevron , though, such a delegation may be 



Text at notes 187-196, infra. 
* Chevron . 467 U.S. at 843-44, quoted at note 61 infra . 



376 ROBERT A. ANTHONY 



presumed from the mere existence of a "gap" — statutory silence 
or ambiguity on the point at issue. More important for present 
purposes, the court must also identify a delegation of power to 
make binding interpretations through the particular format chosen 
by the agency: Did Congress authorize the agency to make law 
through this format? Any difficulty in answering this 
question cannot justify a failure to ask it. That Congress's 
delegatory intent must sometimes be sought through inference and 
construction, rather from direct manifestations of congressional 
will, does not diminish the indispensability of this inquiry. 
Under our system of limited government, an agency cannot announce 
actions that bind citizens and the courts unless Congress (or the 
Constitution) has delegated to it the authority to do so. 

The needed delegation is unlikely to be found for agency 
interpretations expressed through informal formats.^ But, unless 
the courts insist upon a delegation as to format as a condition 
of Chevron acceptance, interpretations set forth in informal 
formats would command as much force as do legislative 
regulations, and agencies could freely avoid the public 
procedures and safeguards required for issuance of such 
regulations. Worse, the courts would accord binding force to 
actions that Congress had not authorized to have binding force. 
Part II of this Article reviews pre- Chevron law concerning 
judicial consideration of agency interpretations. Part III 
analyzes the Chevron doctrine and the experience under it, with 
emphasis on the role of the delegation inquiry. Part IV brings a 
close focus upon the assortment of formats in which agency 
interpretations are set forth. It demonstrates how the delegated 
authority for the use of each format should determine whether the 
reviewing court must accept the agency interpretation or need 
only give it special consideration. 



In the rare case, an informal interpretive format may 
be entitled to judicial acceptance because Congress intended that 
such informal interpretive expressions be authoritative and bind 
the courts if not irrational. See Ford Motor Credit Co. v. 
Mulhollin, 444 U.S. 555, 566-67 (1980). But this apparent 
exception simply proves the rule, since a congressional 
delegation had invested the informally expressed agency 
interpretation with the force of law. See text accompanying note 
153-172, infra . 



AGENCY STATUTORY INTERPRETATIONS 377 

II. PRIOR PATTERNS OF ACCEPTANCE AND DEFERENCE 

No pre- Chevron case articulated a consistent or 
comprehensive statement of doctrine, like that ventured in 
Chevron . on these subjects. Judicial attitudes reflected in the 
opinions range from a near-abject acceptance/" to a skeptical 
consideration of agency views," to an ignoring of them 
altogether.'^ 

Sometimes the judicial language sweepingly called for 
acceptance, but at the same time identified the specific 
circumstances inducing acceptance, leaving doubt about the rule 
of acceptance where such circumstances were not present.'' 

At other times, the opinions spoke not of outright 
acceptance but of degrees of "deference" or "weight" to be 
granted the agency interpretations.'* In these instances the 
"decision whether to grant deference depends on various 
attributes of the agency's legal authority and functions and of 



10 



E.g. . Bates & Guild Co. v. Payne, 194 U.S. 106, 108-9 
(1904), caioted in United States v. Shimer, 367 U.S. 374, 381-82 
(1961). 



" E.g. . Social Security Board v. Nierotko, 327 U.S. 358, 
369 (1946) . 

'^ E.g. . Packard Motor Car Co. v. NLRB, 330 U.S. 469 
(1947) . 

[T]he construction of a statute by those charged with 
its execution should be followed unless there are 
compelling indications that it is wrong, especially 

where Congress has refused to alter the administrative 

construction. 

Red Lion Broadcasting, Inc. v. FCC, 395 U.S. 367, 381 (1969) 
(upholding legislative regulation) . A striking post- Chevron 
example of such a juxtaposition is found in NLRB v. United Food & 
Commercial Workers Union, 108 S.Ct. 413, 421 n. 20 and 
accompanying text (Dec. 14, 1987). 

E.g. ■ Bureau of Alcohol, Tobacco and Firearms v. FLRA, 
464 U.S. 89, 98 n. 8 (1983) ("[A]n agency acting within its 
authority to make policy choices consistent with the 
congressional mandate should receive considerable deference from 
courts * * *") ; Zuber v. Allen, 396 U.S. 168, 192 (1969) ("While 
this court has announced that it will accord great weight to a 
departmental construction of its own enabling legislation, * * * 
it is only one input in the interpretational equation") . 



378 ROBERT A. ANTHONY 



the administrative interpretation at issue." But the pre- 
Chevron cases were habitually unclear in indicating the point at 
which the weight or deference due would compel acceptance of the 
agency interpretation (provided, of course, it was reasonable and 
consistent with statute).^* 

It would not be possible, even if it were fruitful, to posit 
an analysis that harmoniously explained all the pre- Chevron 
cases. But some rather clear patterns are evident. These are 
best perceived in terms of the formats in which the agency 
interpretations were expressed. Where agencies exercised 
delegated power to pronounce interpretations in formats bearing 
the force of law, reviewing courts accepted those 
interpretations, provided they were reasonable and consistent 
with statute. In other formats, the courts accorded special 
consideration to the agency interpretations, but approved them 
only if persuaded that they were correct. 

A. LEGISLATIVE REGULATIONS 

The agency interpretation has regularly been accepted when 

set forth in a legislative regulation issued pursuant to 

delegated authority to promulgate rules with the force of law. 

For example: 

Unlike the statutory term in Title II, however, 
Congress in § 407 (a) expressly delegated to the 
Secretary the power to prescribe standards for 
determining what constitutes "unemployment" for 
purposes of AFDC-UF eligibility. In a situation of 
this kind. Congress entrusts to the Secretary, rather 
than to the courts, the primary responsibility for 
interpreting the statutory term. In exercising that 
responsibility, the Secretary adopts regulations with 
legislative effect. A reviewing court is not free to 
set aside these regulations simply because it would 
have interpreted the statute in a different manner. 
* * * . 



Diver, Statutory Interpretation in th e Administrative 
State . 133 U. Pa. L. Rev. 549, at 56? (19P5). See notes 51-52 
and accompanying text, infra . 

^* Even in an opinion that went to some pains to elucidate 
the issues of "deference," for example, the Court made this 
typical statement: "The interpretation put on the statute by the 
agency charged with administe